Auditory Localisation: Contributions of Sound Location and Semantic Spatial Cues A thesis submitted in fulfilment of the requirements for the degree of Master of Applied Science Submitted by Norikazu Yao B.Sc. (Japan) M.Sc. (Japan) School of Human Movement Studies Queensland University of Technology 2007
108
Embed
Auditory Localisation: Contributions of Sound Location and ...eprints.qut.edu.au/16504/1/Norikazu_Yao_Thesis.pdf · sound is actually used, and whether it assists pure auditory localisation.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Auditory Localisation: Contributions of Sound
Location and Semantic Spatial Cues
A thesis submitted in fulfilment of the requirements for
the degree of
Master of Applied Science
Submitted by
Norikazu Yao
B.Sc. (Japan) M.Sc. (Japan)
School of Human Movement Studies
Queensland University of Technology
2007
i
Keywords
Auditory localization
Spatial Stroop effect
Stimulus response compatibility
Semantic processing
Information processing
Response selection
Reaction time
Orienting
ii
Abstract
In open skill sports and other tasks, decision-making can be as important as
physical performance. Whereas many studies have investigated visual perception
there is little research on auditory perception as one aspect of decision making.
Auditory localisation studies have almost exclusively focussed on underlying
processes, such as interaural time difference and interaural level difference. It is
not known, however, whether semantic spatial information contained in the
sound is actually used, and whether it assists pure auditory localisation. The aim
of this study was to investigate the effect on auditory localisation of spatial
semantic information. In Experiment One, this was explored by measuring whole
body orientation to the words “Left”, “Right”, “Back”, “Front” and “Yes”, as
well as a tone, each presented from left right, front and back locations.
Experiment Two explored the effect of the four spatial semantic words presented
either from their matching locations, or from a position rotated 20 degrees
anticlockwise. In both experiments there were two conditions, with subjects
required to face the position indicated by the sound location, or the meaning of
the word. Movements of the head were recorded in three dimensions with a
Polhemus Fastrak system, and were analysed with a custom program. Ten young
adult volunteers participated in each experiment. Reaction time, movement time,
initial rotation direction, rotation direction at peak velocity, and the accuracy of
the final position were the dependent measures. The results confirmed previous
reports of confusions between front and back locations, that is, errors about the
interaural axis. Unlike previous studies, many more back-to-front than front-to-
back errors was made. The experiments provided some evidence for a spatial
Stroop interference effect, that is, an effect on performance of conflicting
iii
information provided by the irrelevant dimension of the stimulus, but only for
reaction time and initial movement direction, and only in the Word condition.
The results are interpreted using a model of the processes needed to respond to
the stimulus and produce an orienting movement. They suggest that there is an
asymmetric interference effect in which auditory localisation can interfere with
localisation based on semantic content of words, but not the reverse. In addition,
final accuracy was unaffected by any interference, suggesting that these effects
are restricted to the initial stages of response selection.
iv
Table of Contents
KEYWORDS…………………………………………………………………….i
ABSTRACT…………………………………………………………….……….ii
TABLE OF CONTENTS…………………………………………………..…..iv
LIST OF FIGURES......…………………………………………………..……vii
LIST OF TABLES……………………………………………………………....x
LIST OF APPENDICES……………………………………………………….xi
STATEMENT OF ORIGINAL AUTHORSHIP…………………………….xii
LIST OF ABBREVIATIONS………………………………………………...xiii
ACKNOWLEDGEMENTS…………………………………………………..xiv
CHAPTER 1. INTRODUCTION and LITERATURE REVIEW…………...1
1.1 Introduction ………………………………………………………………...1
1.2 Literature Review…………………………………………………………...4
1.2.1. Auditory Localisation………………………………………………….4
1.2.2. Semantic Localisation…………………………………………………8
CHAPTER 2. EXPERIMENT ONE……….…………………………………17
2.1. Introduction..……………………………………………………………...17
2.1.1. Definition of Terms………………………………………………17
2.1.2. Research Question………………………………………………..18
2.1.3. Hypotheses………………………………………………………..19
2.2. Method………... …………………………………………………………..20
2.2.1. Participants...……………………………………………………..20
2.2.2. Experimental environment……………………………………….20
v
2.2.3. Apparatus…………………………………………………………22
2.2.4. Stimulus…………………………………………………………..22
2.2.5. Design and dependent variables………………………………….22
2.2.6. Procedure…………………………………………………………25
2.2.7. Analysis and Statistics ..…….……………………………………25
Responses to the word’s location were particularly slow when the word specified
was in the opposing location on the same dimension. Moreover, Shor (1970) put
the words, LEFT, RIGHT, UP and DOWN into arrows pointing in directions
incongruent with the inserted word and obtained interference in naming the
LEFT
LEFT
Congruent Stimulus Incongruent Stimulus
- 14 -
direction of the arrow. Two task conditions were used, e.g., Arrow task condition:
subjects were instructed to identify (name) the direction of the arrow ignoring the
word. In the Word task condition subjects were instructed to identify (read) the
words, ignoring the surrounding arrows. The results of these studies showed that
RT for naming the direction in which the arrow pointed was slower than naming
the direction word. In addition, naming the direction of the arrow was slower
when the word in the arrow was incongruent, for example, the arrow which
contained the word RIGHT pointing to the left location. However the naming of
the direction word was not significantly slower when the arrow was incongruent.
This suggests that word meaning can interfere with naming a direction that is
indicated symbolically, but not the reverse. In these previous studies, however,
the responses involved naming or saying the location or direction verbally.
Naming and saying the word stimulus were not affected by the incongruent
spatial dimension because these incongruent features or stimuli are not the word
but a shape or location. In other words, the participant has to translate the shape
or location into a word in order to say it, but can use the word directly for a
response to a word stimulus. This leaves it unclear whether the same asymmetry
(the incongruent word slowing performance but not the incongruent arrow)
would be found if spatial, rather than verbal responses were used. In the present
study however, orienting movements were used as the response. Unlike Shor’s
task, in which reading words was not affected by incongruent arrows, this
orienting task could be expected to show more clear-cut interference.
Palef and Nickerson (1978) examined the influence of spatial semantic
information on the speed and accuracy of button-presses, using an auditory
- 15 -
spatial Stroop-like task. They used four spatial word stimuli: “Left”, “Right”,
“Front” and “Back”, and a neutral stimulus “X”, delivered from speakers located
at the Left, Right, Front and Back of the subject. This study was composed of
two conditions, Location and Word. In the Location condition, subjects were
required to respond to the location of the word, by pressing the appropriate
button, while attempting to ignore its meaning. By contrast, in the Word
condition, subjects needed to respond to the meaning of the word stimulus
regardless of its location. As before, when the stimulus dimension corresponded
to the location dimension (e.g., a stimulus “Left” was presented from Left
location), the condition was labelled as congruent. On the other hand, when the
stimulus and location dimensions did not correspond (e.g., a stimulus “Left” was
presented from Right location), this represented an incongruent condition. This
study compared the reaction time for these congruent and incongruent conditions,
using both Location and Word instructions. They found that there were effects of
congruency in the Location but not in the Word condition. Relative to the neutral
condition, these effects included both facilitation for congruent conditions, and
interference, for incongruent conditions. However, these effects were confined
to stimuli presented at the Front and Back locations. The authors concluded that
the results supported the proposal that a necessary condition for the occurrence of
Stroop-like effects is that the irrelevant information is processed faster than, or at
least as fast as, the relevant information. This would explain why they found no
effect of congruence in the Word condition, because they claim that Word
meaning was processed faster than location (at least for Front and Back
conditions). However, this interpretation may be questioned. It relies on
comparing the neutral stimuli in the two conditions. In the Word condition, the
- 16 -
neutral stimulus was a single spatial word (e.g. Left), delivered from all speakers
simultaneously. This is different from the neutral stimulus in the Location
condition.
Another limitation of this study, however, is the fact that it used a key-press
response. A key-press is less closely related to the stimulus, and, in order for the
correct button to be chosen, the task may need more cognitive manipulation than
an orienting movement. An incongruent verbal stimulus in a Word condition (e.g.,
the call of “Left” from the right side) might affect auditory localisation if this
involves an orienting movement. In addition, this study made no mention of
front-back confusions in the results or discussion. If the speaker was positioned
in front and back, specifically on the midline, front-back confusion could be
expected to affect the responses. The present study was concerned with both
front-back confusion and the spatial Stroop effects in orienting movements.
In summary, it is known that humans can localise the source of the pure tone
sound using three different processes that are independent of any semantic
content. However, although it is known that spatial semantic information affects
auditory localization in tasks such as key-pressing, it is unclear if the same is true
for more natural responses, such as whole-body orienting movements. In addition,
measuring the whole movement would provide additional measures than just
reaction time, and these may make clarify the underlying processes.
- 17 -
Chapter 2. EXPERIMENT ONE
2.1. Introduction
The aim of this study was to investigate the effects of spatial and non-spatial
semantic information on sound localization using orienting movements. This
research used methods similar to those of Palef and Nickerson (1978), however,
it differed in several important ways, especially in that a whole body orienting
movement was used, rather than simple button-pressing. Turning to orient to a
sound is a far more typical response and may not make use of the same processes
as those in the button-pressing task, because the sound source directly indicates
the absolute spatial target for orienting, but only indicates the same relative
position for button-pressing. This is because all the buttons were located in front
of the participant in the Palef and Nickerson (1978) study. The latter task may
therefore require additional transformations between stimuli and responses.
2.1.1 Definition of Terms
A non-spatial semantic, or neutral, stimulus is defined as any word that can be
localized through auditory perception, but which has no spatial meaning, such as
the word “Yes”, while a spatial semantic stimulus means a word which has
spatial meaning such as “Left, Right, Front, and Back”. Congruent refers to a
situation in which, for instance, the spatial stimulus “Right” comes from the right
side, or in which the position of a sound source is the same as the semantic
spatial cue (Figure 2.1). In addition, incongruent is defined as a situation in
which the position of the sound source is different from the spatial semantic
stimulus, such as the word “Right” coming from the left, front and back location.
A physical stimulus refers to a non-verbal sound, such as a tone.
- 18 -
Figure 2.1 Left and right congruity between spatial semantic stimulus and position of the sound source: ↑ shows direction in which participant is facing in initial position.
2.1.2. Research Question
Does the spatial semantic content of a word affect the auditory localisation of
that stimulus? By using stimuli with or without spatial semantic content, and by
making the semantic content either congruent, incongruent, or neutral relative to
the actual stimulus location, it was possible to answer this question.
Congruent
Incongruent
Neutral
Congruence Location of spatial semantic stimulus
“Front”
“Back”
“Left” “Right”
“Back” “Left”
“Right”
“Front” “Left”
“Right”
“Right” “Front” “Back”
“Left” “Front” “Back”
“Yes”
“Yes”
“Yes” “Yes”
- 19 -
2.1.3. Hypotheses
2.1.3.1. Hypothesis 1
The speed and accuracy of an orienting movement to verbal stimuli depend on
both auditory localisation of the actual sound and spatial semantic localisation if
it has spatial semantic content. The rationale for this hypothesis is that if spatial
semantic information is not used, then performance will be unaffected by the
addition of spatial semantic content.
The predictions of this hypothesis that were tested are:
1a) Reaction time will be longer when spatial semantic cues and auditory
location cues contained in a stimulus are incongruent rather than neutral
or congruent.
1b) Directional accuracy will be lower when spatial semantic cues and
location cues contained in a stimulus are incongruent rather than neutral
or congruent.
2.1.3.2. Hypothesis 2
Although it was not the primary aim of this study, the experiments offered an
opportunity to test predictions about the occurrence of front-back confusions.
The following hypothesis was formulated:
When auditory location cues and spatial semantic cues are incongruent,
confusion between front and rear occurs more than for left and right. The
rationale for this hypothesis is ambiguity of absolute front and rear space because
of ITD and ILD effects.
Two predictions of this hypothesis were tested:
- 20 -
2a) Reaction time will be longer for front and back than for left and right
positions.
2b) Directional accuracy will be lower for front and back than left and
right positions. Directional judgments will tend to be opposite the front
and back positions (i.e. reversed around the left-right or transverse axis).
2.2. Method
2.2.1. Participants
Ten Queensland University of Technology students aged 18-32 (3 women, 7 men)
who reported no auditory problems were selected as participants. All recruitment
and other procedure were approved by the QUT Human Research Ethics
committee.
2.2.2. Experimental environment
The experiment took place in a semi-reverberant room measuring 9.5m in width,
5.25m in length and 3.40m height. Loudspeakers were placed 2.5m away from
the participant’s standing location in four positions, front, back, left, and right at
a height of 1.5m (Figure 2.2a). A visual surround was positioned between the
participant and the loudspeakers to occlude the participant’s view of the
loudspeakers. The visual surround comprised sections of dark grey cloth draped
from a series of 12 wooden frames, centred on the participant’s standing position.
The visual surround was 2.25m in diameter.
- 21 -
Figure 2.2(a). Layout of apparatus: Overhead view.
Figure 2.2(b) Layout of apparatus: Lateral view.
Visual surround
Loudspeaker
Reference Position
2.5m
Left
Back
Front
Right
Loudspeaker
Reference Mark
Visual Surround
- 22 -
2.2.3. Apparatus
Stimulus presentation and data collection were controlled by a Toshiba Computer,
using a custom program written in the Labview programming language (Version
6.0, National Instruments). The directional error and reaction time were
measured from head movements recorded by the POLHEMUS Fastrak motion
analysis system, which permitted the 3-dimensional position and 3-axis
orientation of a head-mounted sensor to be tracked at 60Hz. A sensor was
mounted on a tight fitting skull-cap worn by the participant.
2.2.4. Stimuli
The verbal stimuli were recorded in a male voice, and were presented through
one of eight loudspeakers. The duration of each stimulus was approximately
500ms and its intensity was 64dB, measured with an industrial sound meter.
2.2.5. Design and dependent variables
The experiment had two main conditions (Location and Word). The form of the
Location condition was a 6 (stimulus type) × 4 (position) pseudo-randomised
design. There were 5 repetitions of each of these 24 conditions. The independent
variables were stimulus type (the words “Left”, “Right”, “Front”, “Back” “Yes”,
and a physical sound (Tone), position (Left, Right, Front, and Back). Front is
aligned with the reference position shown in Figure 2.3. The Word condition was
a 4 (stimulus type; the words “Left”, “Right”, “Front” and “Back”) × 4 (position:
Left, Right, Front and Back) design, with each combination again having five
repetitions. For both conditions, each combination of stimulus and location was
- 23 -
used once before any repetition occurred, making 5 consecutive blocks of trials,
each containing all combinations.
2.2.5.1. Reaction time
Reaction time was defined as the interval between the sound being presented and
the participant starting to move. This was obtained in post-collection data
analysis with a computer program algorithm which identified the first sample
changing by more than 0.5 degrees in rotation or inclination, the lowest practical
threshold based on pilot testing. A graphical display of the kinematics of each
movement allowed visual checking of the point at which the movement was
identified as starting.
2.2.5.2. Initial rotation direction
The initial rotation direction (IRD) was determined as the direction of the
movement (clockwise or anticlockwise) in the first 100 ms following movement
onset. It was expressed as an index by calculating the frequency of clockwise
movements across repetitions within subject, such that 1 = 100% clockwise
movements and 0 = 100% anticlockwise movements.
2.2.5.3. Rotation direction at Peak Velocity
The direction of motion at the time of peak rotation velocity (RDPV) was
analysed in the same way as IRD.
2.2.5.4. Movement time
The Movement time was determined as the time from the initial movement to the
end of the movement.
2.2.5.5. Constant error
Directional error (or Constat error, CE) was defined as the difference between
the target and judgment positions, and was indicated as positive (+nº) for errors
- 24 -
in a clockwise direction, and negative error (-nº) for a anti-clockwise direction.
Figure 2.3 shows this directional error calculation. If the participant judged the
target sound source to be in position A (Figure 2.3) relative to the actual (target)
sound source, the error from location B (Figure 2.3) would be -18º.
2.2.5.6. Reliability
The consistency of directional error was calculated as Reliability (R). Reliability
was calculated through the use of circular statistics, in which highly consistent
performance approaches a value of one, and inconsistent performance
approaches zero. The actual value is the length of the resultant vector of all the
repetitions within a condition for each subject. This measure is preferred to
Variable Error in situations where errors of, for example, -179 degrees and +179
degrees can occur in the same condition, producing very high VE values even
though they differ by only two degrees. Reaction time was measured using data
from the head-mounted sensor.
Figure 2.3. Directional error calculation illustrated for a “Back” target. A and B depict two possible head orientations relative to the target shown by the loudspeaker (L).
B A
L
+10º -18º
Back
- 25 -
2.2.6. Procedure
The participant was instructed in the task and, at the beginning of each trial,
stood facing a reference mark at eye level in the front-facing initial position. The
stimuli were presented in a random order by the laptop computer. Participants
turned to face the sound source as quickly and accurately as possible and
remained still in the Location condition, whereas participants turned and faced
the location indicated by the word stimulus in the Word condition. The order of
the Word and Location conditions was counterbalanced across participants.
Participants were instructed to face squarely the location appropriate for each
condition as instructed by the experimenter. Participants had to turn their whole
body towards that location. Special instructions were given for cases where the
participant judged the relevant position to be straight ahead. Since a movement
of some type was required in order for reaction time to be measured, the
participant was asked to nod his or her head if he or she judged that the sound
source was straight ahead. Participants could make corrective movements within
the 2.5s sampling period. The 3D position and orientation of the participant’s
head was recorded throughout each trial. This position was referenced to the
straight ahead (Front) position. Participants were given a short break (three
minutes) between blocks of trials. A set of five practice trials was presented first.
Blocks of localisation trials followed, each composed of 24 random presentations
of stimuli through each of the loudspeaker locations.
2.2.7. Analysis and statistics
The “Yes” and tone stimuli in the Location condition were first compared with
the other stimuli in a 6 × 4 repeated measure analysis of variance. Since there
- 26 -
were “Yes” and tone stimuli only in the Location condition, the main analyses
were then conducted using only the four semantic spatial stimuli in a 2 (location
vs. word instruction) × 4 (stimulus location) × 4 (stimulus meaning) repeated
measures ANOVA. In addition to RT and Directional Error, the following
dependent variables were obtained: the proportion of clockwise and anti-
clockwise movements, the initial direction of movement, peak rotational velocity
and movement time. Planned comparison analyses were performed for specific
pairwise comparisons, and Fisher’s post-hoc test was used to further examine
significant effects from the ANOVAs. In all figures, error bars represent 95%
confidence intervals.
2.3. Results
2.3.1. Reaction Time (RT)
The first analysis compared, in the Location condition, the two “control” stimuli
(“Yes” and Tone), with the four stimuli that have semantic spatial content
(“Left”, “Right”, “Front” and “Back”). The word “Yes” acts as a control for
semantic spatial content, since it is verbal but its meaning does not refer to any
location. The tone stimulus has no semantic content of any kind. The reaction
times (RTs) for these two control stimuli did not differ significantly from the
spatial semantic stimuli, F(5,45)=.45, p=.81, nor was there any interaction
between Stimulus and Location, F(15,135)=.73, p=.74. These data are shown in
Figure 2.4. Subjects took longer to start their orientation movements when
instructed to use the location of the stimulus, rather than its meaning. The overall
mean reaction time for the Location condition (619 ±135ms) was significantly
longer than for the Word condition (539 ±133ms), F (1,9) =5.33, p<.05. This is
- 27 -
shown in Figure 2.5, which also depicts the RT for each stimulus (the word
“Left”, “Right”, “Front” and “Back”).
"Left" "Right" "Front" "Back" "Yes" Tone
STIMULUS
200
300
400
500
600
700
800
900
1000
1100R
eact
ion
Tim
e (m
s) LOC Left LOC Right LOC Front LOC Back
Figure 2.4 Reaction time for each stimulus in the Location condition (Points represent mean values; errors bar in all figures represent 95% confidence intervals). Values for each location are joined by lines.
- 28 -
"Left" "Right" "Front" "Back"STIMULUS
200
300
400
500
600
700
800
900
1000
1100
Rea
ctio
n Ti
me
(ms)
Location Condition Word Condition
Figure 2.5 Reaction Time for each stimulus in the Location and Word conditions. The “Front” stimulus in the Word condition had significantly longer RTs than the
other three stimuli, F(2,27)=31.47,p<.00001). In addition, post-hoc analysis
showed that reaction times for each location were similar in the Word condition,
whereas RTs for front and back positions were both significantly longer than for
left and right positions (p<.01). These values are shown in Figure 2.6. A specific
comparison was made of RTs for congruent and incongruent conditions as a test
of Hypothesis 1a. This stated that “Reaction time will be longer when semantic
spatial cues and auditory location cues contained in a stimulus are incongruent
rather than neutral or congruent”. This involved comparing RTs for the word
“Left” delivered from the left (or “right” from the right), i.e. congruent
conditions, and those for the word “Left” delivered from the right (or “right”
from the left), i.e. incongruent conditions.
- 29 -
Left Right Front BackSTIMULUS LOCATION
200
300
400
500
600
700
800
900
1000
1100
Rea
ctio
n Ti
me
(ms)
Location Condition Word Condition
Figure 2.6 Reaction time for each location (Left, Right, Front and Back) in the Location and Word conditions.
These RTs are shown in Figure 2.7. There was a strong effect of congruence in
the Word condition with congruent RTs about 76ms shorter for congruent
conditions. This pattern of RT data confirmed hypothesis 1a, but only for the
Word condition. This was the only “pure” analysis of the spatial Stroop effect in
this experiment, as any comparisons involving Front and Back locations are,
potentially, affected by Front-Back confusions and, for the Front locations, the
unique requirement of nodding the head rather than rotating.
- 30 -
LOCATION CONDITION
Congruent Incongruent300
350
400
450
500
550
600
650
700
Rea
ctio
n Ti
me
(ms)
WORD CONDITION
Congruent Incongruent
STIM-LOC Left STIM-LOC RIght
Figure 2.7 Reaction Time for congruent and incongruent stimulus-location pairs (“Left” from left and right side, and “Right” coming from left and right side)
For the Location condition only, and for just the Left and Right positions, it was
possible to compare the congruent and incongruent RT values with those for the
neutral stimuli (“Yes” and Tone). The results were very similar for comparisons
with both the “Yes” and Tone stimuli. The neutral stimuli were longer than
either congruent or incongruent values (averaged over left and Right locations):
values can be obtained from the appropriate parts of Figure 2.4). To examine
whether there was a difference between congruent, incongruent and neutral
stimuli, a separate 3 (congruence) x 2 (stimulus location: left or right) ANOVA
was run, and a planned comparison of the neutral stimulus against congruent and
incongruent values was calculated. This analysis was undertaken twice, once
using the “Yes” stimulus, and once using the Tone stimulus as the neutral
- 31 -
condition. Comparisons involving the “Yes” stimulus showed no main effect of
congruence (F (2,18)=2.13, p=0.15), nor a significant planned comparison effect.
(F (1,9)=2.60, p=.14). Comparisons with the Tone stimulus showed that it had a
significantly longer RT (main effect of congruence: F (2,18)=5.24, p<.05;
planned comparison, F (1,9)= 6.70, p<.05).
2.3.2. Initial Rotation Direction.
In Figure 2.8, it can be seen that subjects almost always rotated anticlockwise for
Left locations and clockwise for Right locations in the Location condition, while
in the Word condition, they usually rotated anticlockwise for the “Left” stimulus
and clockwise for the “Right” stimulus. In both conditions, clockwise and
anticlockwise directions were chosen about equally often for Front and Back
locations (Location condition), and “Front” and “Back” stimuli (Word condition).
These results, overall, show that subjects generally rotated in the direction
required for the instructions they had been given. However, there was clear
evidence that, in the Location condition, subjects were not influenced by word,
but in the Word condition, they were influenced by location. This is clear in
Figure 2.8 (right panel). Regardless of the stimulus, subjects were significantly
more likely to turn in an anticlockwise direction if the location of the stimulus
was on the left than if it was on the right, planned comparison: F(1,9) = 6.08,
p<.05.
- 32 -
LOCATION CONDITION
LOC:Left
RightFront
Back-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Initi
al R
otat
ion
Dire
ctio
n
WORD CONDITION
LOC:Left
RightFront
Back
STIM "Left" STIM "Right" STIM "Front" STIM "Back"
Figure 2.8 Initial Rotation Direction in the Location and Word conditions. (0 and 1 indicate 100% frequency of anti-clockwise and clockwise movements, respectively.)
2.3.3. Rotation Direction at Peak Velocity
The patterns of Rotation Direction at Peak Velocity are similar to those for IRD
with one important difference (Figure 2.9). In the Word condition, the difference
between Left and Right locations described above for IRD are much smaller for
the “Front” and “Back” stimuli, and not present at all for the “Left” and “Right”
stimuli. The Stroop interference shown for IRD in the Word condition is much
reduced or absent by the time of peak velocity.
- 33 -
LOCATION CONDITION
LOC:Left
RightFront
Back-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Rot
atio
n D
irect
ion
at P
eak
Vel
ocity
WORD CONDITION
LOC:Left
RightFront
Back
STIM "Left" STIM "Right" STIM "Front" STIM "Back"
Figure 2.9 Rotation Direction at Peak Velocity in the Location and Word conditions 2.3.4. Movement Time
No significant difference for movement time between the stimuli was found in
the Location condition, while movement time for the “Back” stimulus was
significantly longer than any of the other stimuli in the Word condition, F(3,
27)=11.79, p<.00005, Figure 2.10. Movement time did not differ significantly
either between conditions (Word vs. Location) or between the four locations
(Figure 2.11).
- 34 -
"Left" "Right" "Front" "Back"
STIMULUS
400
600
800
1000
1200
1400
1600
1800
Mov
emen
t Tim
e (m
s)
Location Condition Word Condition
Figure 2.10 Movement Time for each stimulus in the Word and Location conditions
Left Right Front Back
STIMULUS LOCATION
400
600
800
1000
1200
1400
1600
1800
2000
2200
Mov
emen
t Tim
e (m
s)
Location Condition Word Condition
Figure 2.11 Movement Time for each location in the Word and Location conditions
- 35 -
2.3.5. Front-Back Reversals
Table 2.1 and 2.2 shows the number and percentage of front-back reversals. The
only factor affecting these errors was the location of the stimulus, with the errors
occurring overwhelmingly for the back positions, i.e. back-front reversals. This
pattern confirms hypothesis 2b, which predicted more front-back than left-right
reversals. However, the hypothesis was confirmed only for the Location
condition.
Table 2.1. Frequency of front-back reversals in the Location condition (number and percentage of trial in each combination of location and stimulus). Stimulus "Left" "Right" Location Left Right Front Back Left Right Front Back Number 0 0 1 25 0 1 2 30 % 0 0 2 50 0 2 4 60 Stimulus "Front" "Back" Location Left Right Front Back Left Right Front Back Number 0 0 1 34 0 0 3 30 % 0 0 2 68 0 0 6 60 Stimulus "Yes" Tone Location Left Right Front Back Left Right Front Back Number 0 0 0 31 0 0 0 27 % 0 0 0 62 0 0 0 54
Table 2.2 Frequency of front-back reversals in the Word condition (number and percentage of trial in each combination of location and stimulus). Stimulus "Left" "Right" Location Left Right Front Back Left Right Front Back Number 0 0 0 0 0 0 0 0 % 0 0 0 0 0 0 0 0 Stimulus "Front" "Back" Location Left Right Front Back Left Right Front Back Number 0 0 0 0 2 0 0 2 % 0 0 0 0 4 0 0 4
- 36 -
2.3.6. Constant Error
The directional error at the end of each movement was assessed through Constant
Error (CE). Front-back reversal errors were corrected according to the method of
Oldfield (1986). This involved “mirroring” the final position around the
participants’ inter-aural axis, and using the new value if this was smaller than the
original value. In effect, this procedure removes the front back reversal
component of the error. Overall, there were no systematic effects of conditions
on CE. Figure 2.12 shows values for Constant Error for combination of condition,
location, and stimulus. However, a wide range of errors was found for each of
the back locations in the Location condition, as seen in the large errors bars in
Figure 2.12. Note that in the Word condition, CE was very close to zero and
quite consistent, except for the “Right” stimulus, which had slightly larger errors.
LOCATION CONDITION
LOC:Left
RightFront
Back-6
-4
-2
0
2
4
6
8
10
12
14
16
Con
stan
t Err
or (d
egre
e)
WORD CONDITION
LOC:Left
RightFront
Back
STIM "Left" STIM "Right" STIM "Front" STIM "Back"
Figure 2.12 Constant Error for each stimulus and each location in the Location and Word conditions
- 37 -
Even though CE was lower for congruent than for incongruent combinations in
the Location condition, (Left and Right stimuli and locations only) the difference
was not significant, planned comparison, F(1,9)=1.24 p=.30. Therefore,
Hypothesis 1b, which stated that errors would be lower for congruent conditions,
was not confirmed. (Values for conditions including the “Yes” and “Tone”
stimuli are shown in Appendix H and I. These confirm that accuracy did not
differ between the spatial and control stimuli).
2.3.7. Reliability
These values of Reliability (R) are shown in Figure 2.13. The Back location had
lower reliability (i.e. more trial-to-trial variability) than the other stimuli, but
only in the Location condition. This is shown in Figure 2.13, but could not be
formally analysed because of the extreme variations in variances for the four
stimuli.
Left Right Front BackLOCATION
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
Rel
iabi
lity
Location Condition Word Condition
Figure 2.13 Reliability for each location in the Location and Word conditions
- 38 -
2.4. Discussion
The actual findings with regard to the spatial Stroop effect are explained below
by considering the results of each dependent variable, presented in chronological
order. Then the effect of Front-Back confusion on the auditory localization is
discussed with some variables in this experiment, even though the Front-Back
confusion is not directly related to the aim of this study. It should be noted that it
was not possible to present the Stroop data for RT as difference scores relative to
a neutral condition. This was because no neutral condition could be used in the
Word condition, and because, in the Location condition, any comparisons
involving Front and Back locations are potentially affected by the occurrence of
Front-Back confusions, and by the special case of a nodding movement for the
Front location. Thus absolute RT values only were examined.
2.4.1. Spatial Stroop Effect
This study investigated the effects of the spatial semantic cues for a localisation
task by testing for the presence of a spatial Stroop effect. The hypotheses were
also based on previous accounts of the spatial Stroop effects. It was assumed that
spatial semantic cues reduce reaction time and increase accuracy of the response,
if the relevant and irrelevant stimulus dimensions are congruent (Stimulus “Left”
is uttered from left location).
2.4.1.1. Reaction Time and Initial Rotation Direction
The Spatial Stroop effect was apparent in the Left and Right dimension of the
orienting movement in the Word condition. Reaction time for the congruent
condition (e.g., “Left” from the left location) of an orienting movement to the
- 39 -
word instructions was significantly faster than in incongruent conditions (e.g.,
“Left” from the right location). In the Location condition, however, the spatial
semantic cues did not influence the RTs at all. In the Word condition, the
orienting movement for word instructions was influenced by the location
dimension, but the semantic stimulus dimension did not affect the Location
condition. The spatial Stroop effect therefore was asymmetrical, occurring only
when subjects were attempting to use only the semantic content of the stimulus
word, and only in some conditions. The reason why the spatial semantic cues did
not affect the Front and Back location is most likely that front-back reversal
errors occurred more frequently in the Location condition, causing RTs, to be
longer. In addition, the Front movement in this study was a nod, i.e. a different
movement from the other locations. This may have affected the RT for this
location only. RT for the front movement was significantly longer than the other
locations. This pattern of results suggests that auditory localisation may influence
semantic processing, but not the other way round.
Neutral stimuli had longer RTs than either congruent or incongruent stimuli, by
about 45 ms, when comparing just the left and right locations (which were not
affected by the unique case of the nodding movement). While the “Yes”
stimulus was not significantly longer, values for Tone were. This outcome is not
consistent with either clear-cut interference or facilitation effects, as the
incongruent values might be expected to be longer than neutral values. It may be
that the relatively low frequency of these neutral stimuli compared to semantic
spatial stimuli (1/6 of the trials in each case) caused participants to respond more
- 40 -
slowly simply because they were unusual. A proper assessment of the neutral
stimuli would require that they are delivered with equal probability.
The initial rotation direction (IRD) for all stimuli in the Word condition, but most
clearly that for the “Back” stimulus, also showed a clear influence of location
processing (i.e. processing of the irrelevant stimulus dimension) very strongly.
The mean IRD expresses the direction which subjects tend to begin their
orienting movement - either clockwise or anti-clockwise at the start of the
movement. Subjects more often started to turn anti-clockwise when the stimulus
was presented from the Left location, and more often clockwise when it was
from the Right location. The subject’s initial direction seemed to be affected by
auditory localisation processing in the Word condition. A possible cause for these
results is that subjects cannot easily prevent themselves starting to respond to the
stimulus as a sound. If auditory localisation processing commences faster than
spatial semantic processing, and it would influence response selection process
strongly. Therefore, the interference of the location content in the Word
condition affected RTs and IRD because they are in the initial part of the
orienting movement. As the other possible cause of interference, an orienting
movement can be assumed to be a natural consequence of sound localisation,
namely, there is an innate link between the auditory system and orienting to the
source of a sound regardless of the task (e.g., the Location or Word condition).
2.4.1.2. Rotation Direction at Peak Velocity, Movement Time and Constant Error
There is no difference between the congruent and incongruent condition in
Experiment One for any of these measures. In addition, no difference between
- 41 -
the congruent, incongruent, neutral and physical tone was found in the Location
condition. Both auditory localisation and spatial semantic processing would
generally be complete before the end of response selection and the subsequent
orienting movement. Previous studies that investigated the Simon and Stroop
effects used key-pressing or joystick estimation as the response. These responses
do not include adjustment of the final location, pressing a correct key or making
a discrete movement of joystick does not include continuous feedback processing
or corrections. The Simon and Stroop-like effects would not affect the whole
response selection but simply the first part of the process. Because the current
task involved a movement lasting approximately 900-1400ms, there was plenty
of opportunity for modification and correction of the movement. As compared
with IRD, which showed interference by sound location in the Word condition,
RDPV showed no such effect. This suggests that interference effects were
present only early on, and by the time of peak movement velocity, only the
relevant stimulus was being used.
The findings of this study contrast strongly with the one previous study which
used a similar experimental design (Palef & Nickerson, 1978). Palef &
Nickerson (1978) reported spatial Stroop effects in the Location but not the Word
condition, whereas in the current study, they were found in the Word condition
but not the Location condition. It was not clear what these different results were
caused by in the two studies. One possible cause is that the response to the
stimulus in the current study was an orienting movement (turning the whole
body). As previously stated, an orienting movement may be more natural and
have a faster RT than would a keypressing response. The keypressing would be
- 42 -
slightly more complicated for the response task because the subjects should have
to decide which button to press using more cognitive rules. The difference
between the present results and those of Palef and Nickerson may also be
understood by examining the time course of processing for the semantic and
location aspects of the stimuli. This is explored more fully in the General
Discussion. This current experiment provides new information about the time
course of the spatial Stroop effect. It was evident only in the RT and initial
rotation direction, and had disappeared by the time of peak rotation velocity. This
suggests that spatial Stroop effects can be corrected after just a few hundred
milliseconds.
2.4.2. Front Back Confusions
2.4.2.1. Reaction time
In this experiment, reaction time for left and right locations was significantly
faster than for front and back locations in the Location condition. There are well
accepted explanations for front back reversals from previous studies. First, a
large number of studies (Abel, Figueiredo, Consoli, Birt, & Papsin, 2002;
1979) showed that front-back reversals and confusion were accounted for by the
“cone of confusion” which refer to a cone-shaped zone in which subjects often
mistake front locations for back, and vice versa. These studies suggested that the
cone of confusion is based on Interaural Time difference (ITD) and Interaural
Level Difference (ILD). As was described in Chapter 2, interaural differences
- 43 -
are smallest when the sound comes from the midline (i.e. immediately in front or
behind - on the mid-sagittal axis). Wallace and Fisher (1998) indicated in their
sound localisation study that response time for keypresses increased when sounds
were presented from speakers symmetrically located in front of and behind the
participant. Moreover, they indicated that front-back confusion made response
times longer, regardless of whether the speakers were positioned on the medial or
lateral axis. These findings are very similar to those in the current experiment.
2.4.2.2. Constant Error and Reliability
The expectation of this study was that directional accuracy would be lower for
front and back than for left and right locations. A large number of front-back
reversal errors were found in this experiment. Constant Error was calculated by
correcting for front-back reversal errors. The reliability measure indicated high
levels of variability, however. A large range for Reliability was found in the
Location condition. Gilkey & Anderson (1995) reported front-back reversals in a
sound localisation task using spoken word stimuli. In their study, localisation
accuracy for front and back locations was lower than for Left and Right locations.
A most interesting outcome in the current study was that the rate of front-back
confusion was different from previous studies: 2% from front to back reversals
and 59% from back to front reversals. Begault & Wenzel, (1993) reported,
however, the percentage of reversals from front to back was significantly higher
than from back to front (47% v 11%), using speech stimuli from eight locations
with an oral response (and for midline stimuli only, front-back reversals (0
degree, 180 degree), front-back reversals: 58% and back-front reversals: 24%).
In addition, Wightman and Kistlaer (1989b) reported 29% front-back reversal
and 6% back-front reversals in the middle elevations. A possible reason for this
- 44 -
difference is the method of presentation of the stimuli. They presented stimuli
over headphones based on filtering by head-related transfer functions (HRTFs),
methods which have recently been developed to synthesize a virtual sound
source. However, a difference between the studies based on HRTFs and free field
sound localisation studies may be shown by this current study. The method
should be considered in future studies for investigating front-back confusion.
However, it is also possible that the acoustic properties of the testing room in the
present study were asymmetric, and if sounds from behind echoed from the front
more often than the reverse, then this could explain the findings.
- 45 -
Chapter 3 Experiment Two
3.1 Introduction
In order to obtain a separate estimate of the influence of semantic information, a
second experiment was undertaken. In previous studies, the incongruent
condition was very obviously incongruent, for example, the word “Left” was
presented from Right, Front and Back positions, similar to Experiment One in
this study. In such cases, there is a 90º or180º difference between the two
dimensions of the stimulus. It could be that increased RTs for an incongruent
condition (as was found for the Word condition in Experiment One, for left and
right stimuli) only occur when the incongruence is very obvious. If the
difference between the sound source location and the semantic spatial content is
much less clear-cut, and an advantage for congruence is still found, this would be
strong support for the idea that semantic content is obligatorily processed as part
of response selection. Therefore, this experiment examined performance when
the degree of incongruence was much smaller than in Experiment One.
3.1.1. Hypothesis
The auditory localisation process uses semantic spatial localisation cues in
addition to auditory localisation cues.
Two predictions of this hypothesis that were tested are:
1. Reaction time will be longer when the sound source is rotated from the
front, back, left and right positions by 20 degrees to the left (anti-
clockwise, from above).
2. Directional estimates will be biased towards the position indicated by
the semantic cues under rotated conditions than non-rotated conditions.
- 46 -
3.2 Methods
Apparatus and stimuli were identical with, and the general procedure was similar
to Experiment 1. However, a new group of participants aged 18-23 (5 women, 5
men) was recruited.
3.2.1. Experimental environment
Participants performed ten repetitions in the same task as in the first experiment.
The major differences were that:
• Loudspeakers were placed in four positions as in Experiment 1, but
there were two conditions: non-rotated and rotated (anticlockwise by
20 degree, Figure 3.1).
• Only the four congruent stimuli (Front, Back, Left and Right) were
used. (Note: although these are described as ‘congruent’ in this case,
in the sense that, for example, a “Front” word stimulus was always
presented from a position closer to the actual front than to the left,
right or back positions, there is still a form of incongruence when the
word meaning denotes a position 20º rotated from the location of the
sound source).
3.2.2. Design
The experiment had a similar design to Experiment 1, however, the form was a 2
(rotation) × 4 (position) design. The independent variables are rotation (absolute
position and rotated position), and directional position (Front, Back, Left and
Right). Dependent variables are the same as Experiment 1.
- 47 -
Figure 3.1 Layout of apparatus for Experiment Two. Speakers were positioned 20 degrees anti-clockwise in the Rotated condition.
3.2.3. Procedure
The procedure of Experiment Two was identical to Experiment One. The rotated
and non-rotated conditions were randomised within the separate Location and
Word conditions which were counterbalanced.
3.2.4. Analysis and statistics
The same procedures for identifying RT and directional error were used as in
Experiment 1. Since there were no tone or “Yes” stimuli, all analyses used a 2 ×
4 × 4 repeated measure ANOVA, with data averaged over repetitions. Planned
comparison analyses were performed for specific pairwise comparisons, and
Fisher’s post-hoc test was used to further examine significant effects from the
ANOVAs.
Speaker
Reference Position
Non-rotated Rotated
20º
- 48 -
3.3. Results
3.3.1. Reaction Time
Reaction time for the Word condition (469±92ms ) was significantly shorter (by
146 ms) than the Location condition, 615±174ms, F(1,9)=16.76, p<.005 (Figure
3.2). Hypothesis 1a predicted that RTs would be longer under rotated than non-
rotated conditions. This hypothesis was not confirmed, however. In fact, for the
Front location in the Location condition, the reverse was true (planned
comparison F(1,9)=22.74, p<.005). There was hardly any effect of rotation in the
Word condition. In addition, the RTs for the stimuli ”Left” and “Right” were
significantly faster than for the stimuli “Front” and “Back” in both Location
F(1,9)=32.77, p<.0005 and the Word conditions F(1,9)= 28.09, p<.0005).
LOCATION CONDITION
STIM-LOC:Left
RightFront
Back200
300
400
500
600
700
800
900
1000
1100
Rea
ctio
n Ti
me
(ms)
WORD CONDITION
STIM-LOC:Left
RightFront
Back
Non-ROTATION ROTATION
Non-ROTATION ROTATION
Figure 3.2 Reaction Time for the four stimuli in the Word condition and for the four locations in the Location condition (STIM-LOC denotes stimulus and location, which are always congruent in Experiment Two.)
- 49 -
3.3.2. Initial Rotation Direction
Initial rotation direction for the “Front” stimulus in the rotated condition was
significantly more often anticlockwise than for the non-rotated condition
(planned comparison, F(1,9)=29.45, p<.0005) and the “Back” stimulus in the
rotated condition was significantly more often clockwise than the non-rotated
condition in the Location condition, planned comparison F(1,9)=16.58, p<.005,
Figure 3.3. This effect was not found in the Word condition.
LOCATION CONDITION
STIM-LOC:LEFT
RIGHTFRONT
BACK-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Initi
al R
otat
ion
Dire
ctio
n
WORD CONDITION
STIM-LOC:LEFT
RIGHTFRONT
BACK
Non-ROTATION ROTATION
Figure 3.3 Initial Rotation Direction for each stimulus for the Non-Rotated and Rotated conditions 3.3.3. Movement Time Figure 3.4 depicts movement time for each stimulus or location in each condition.
Movement time for the Back location in the Location condition was significantly
longer than the other locations, F(3, 27)=3.91, p<.05. In addition, Movement
time for the “Back” stimulus in the Word Condition was significantly longer than
the other stimuli, F(3, 27)=9.24, p<.0005.
- 50 -
LOCATION CONDITION
STIM-LOC:LEFT
RIGHTFRONT
BACK600
800
1000
1200
1400
1600
1800
2000
Mov
emen
t Tim
e (m
s)
WORD CONDITION
STIM-LOC:LEFT
RIGHTFRONT
BACK
Non-ROTATION ROTATION
Non-ROTATION ROTATION
Figure 3.4 Movement Time for each stimulus or location in both conditions
3.3.4. Front-Back Reversals
Front back reversal errors were also found for the back location in Experiment
Two (Table 3.1), however the front back reversal errors were not as frequent as
in Experiment One.
3.3.5. Constant Error
As in Experiment One, front-back reversals were corrected using the method of
Oldfield & Parker (1986) for the analysis of Constant Error. Constant error for
the Rotation condition was significantly higher than for the Non-Rotation
condition, F(3, 27)=3.04, p<.05 (Figure 3.5).
Table 3.1 Frequency of front-back reversals in the Non-rotated and rotated condition (number and percentage of trial in each combination of location and stimulus). CONDITION Non-Rotation Rotation LOCATION Left Right Front Back Left Right Front Back Frequency 1 0 3 21 0 0 0 11 % 1 0 3 21 0 0 0 11
- 51 -
LOCATION CONDITION
STIM-LOC:LEFT
RIGHTFRONT
BACK-30
-25
-20
-15
-10
-5
0
5
10
15
20
25
Con
stan
t Err
or (d
egre
e)
WORD CONDITON
STIM-LOC:LEFT
RIGHTFRONT
BACK
Non-ROTATION ROTATION
Non-ROTATION ROTATION
Figure 3.5 Constant Error for each stimulus in both the Location and Word condition. The main results for CE are shown in Figure 3.5, 3.6, 3.7. In the second two
figures, 3.6 and 3.7 results are shown in the form of arrows, pointing in the
direction of the average CE for each condition. The solid arrows in Figure 3.6
show that the non-rotated condition was reasonably accurate (although subjects
tended to turn about 7 degree clockwise from the stimulus in the Back condition).
When the rotated condition is compared to the non-rotated condition, it can be
seen that, for Front and Back locations, subjects rotated by close to the 20 degree
offset, in the appropriate direction. The rotation in the Left and Right conditions,
however, was much smaller (approximately 5 degrees). The interaction between
rotation and location was significant, F(1,9)=5.23, p<.05.
- 52 -
Figure 3.6 Constant Error for Non-rotation and Rotation condition in the Location condition
Figure 3.7 Constant Error for Non-rotation and Rotation condition in the Word condition
Rotated Right
Non-Rotated Back
Non-Rotated Right
Non-Rotated
Left
Rotated Front
Rotated Left
Rotated Back
Non-Rotated Front
Non-Rotated Direction judgement
Rotated DirectionSD Non-Rotated
SD Rotated
Rotated Right
Rotated Front
Rotated Left
Rotated Back
Non-Rotated Front
Non-Rotated Right
Non-Rotated
Left
Non-Rotated Back
Non-Rotated Direction judgement
Rotated DirectionSD Non-Rotated
SD Rotated
- 53 -
Figure 3.7 (Word Condition) shows that there was little if any effect of rotation.
Subjects pointed quite accurately to the position indicated by the stimulus, i.e. to
the Left (270 degree), Right (90 degree), Front (0 degree) and Back (180 degree).
3.3.5. Reliability
Although extreme differences in variance make a formal statistical comparison
less reliable, Reliability for the Location condition was significant lower than the
Word condition, F(1, 9)=9.51, p<.05. Moreover, Reliability for the Back location
was significantly lower than for the other locations in the Location condition,
(Fisherman LSD post-hoc analyses: for Left location p<.001, for Right location,
p<.001, for Front location p<.005), whereas Reliability was little changed in the
Word condition (Figure 3.8).
LEFT RIGHT FRONT BACKSTIMULUS-LOCATION
0.70
0.75
0.80
0.85
0.90
0.95
1.00
1.05
1.10
Rel
iabi
lity
Location cond Word cond
Figure 3.8 Reliability of Constant Error in the Location and Word condition
- 54 -
Even when the statistical analysis is disregarded, the extremely consistent
performance in the Word condition, and the more variable performance in the
Location condition for the Back stimulus and location, are very apparent.
3.4. Discussion
In Experiment Two, the key comparison was between non-rotated and rotated
conditions. Unlike Experiment One, there were no cases in which the location
indicated by the stimulus word was completely incongruent with the sound
location. For example, the stimulus “Left” was presented from the non-rotated
location (straight ahead), and from the (rotated) location 20 degree anti-
clockwise. Likewise, the rotated “Left” was not exactly Left but still on the left
side as compared with the other locations. The rotated condition was not
absolutely incongruent, especially with respect to the direction of movement. In
this experiment, in neither the Word nor the Location condition was there any
evidence of a spatial Stroop effect on RT, IRD, RDPV, or MT. A possible reason
for this is that the difference between the positions indicated by the relevant and
the irrelevant dimensions was too small to cause interference. The locations
where the rotated stimuli were presented in both conditions were on the same
side as of the non-rotated condition, (except Front). At one level, the word’s
spatial meaning never really clashed with the actual sound location, while at a
more detailed level, the two locations were in fact slightly different. The two
earliest indications of performance, RT and IRD, might be expected to show
identical effects, but there were actually some differences between the two
measures. There was no difference in RT between the rotated and non-rotated
conditions, in either Word and Location conditions. On the other hand, for the
- 55 -
Front and Back stimuli, there was an extremely clear effect of rotating the
stimulus anti-clockwise. Subjects tended to rotate anti-clockwise much more
often when the sound source was rotated. This shows they were influenced by the
actual location at an early stage. In the Word condition, however, the IRD did not
change. Subjects could ignore location when instructed to do so in this
experiment even for the initial direction, because the directions were not
incongruent.
Similarly, RTs for left and right locations were significantly longer than front
and back locations, and there were also some Front-back reversals in front and
back locations. In addition, this experiment’s data also supports a finding (Abel
& Paik, 2004) that rotated front and rotated back locations had significantly
longer reaction times than both rotated and non-rotated left and right locations.
Compared to these early variables, CE for the Location condition shows a
difference between the rotated and non-rotated conditions clearly. In all four
locations in the Location condition, subjects ended up facing in a more anti-
clockwise position in the rotated conditions. By contrast, they were completely
unaffected by rotation in the Word condition, however. In addition to this
rotation, the CE for rotated Left and Right locations in the Location condition
was biased towards the non-rotated location, i.e. subjects judged the location
towards the position indicated by the word meaning. However, previous sound
localisation studies reported the judgment of left and right locations was
estimated near the interaural axis. Begault & Wenzel, (1993) reported that five
subjects out of ten showed a biased response in which they tended to “pull”
- 56 -
toward the interaural lateral axis. This pattern was also observed in a free-field
study (Oldfield & Parker, 1984). Therefore, it was not clear that in the current
study this bias towards the word location represented a Stroop interference effect
(i.e. an interference of the word meaning) or whether it was a simple bias
towards the interaural lateral axis that is independent of the stimulus properties.
- 57 -
Chapter 4. GENERAL DISCUSSION The aim of this study was to investigate the effects of spatial semantic
information on sound localisation using orienting movements. In order to
examine this, it was hypothesised that reaction time (RTs) for, and constant error
(CE) of location judgement would be affected by spatial semantic information if
this information and the actual sound location are incongruent, in other words,
that a spatial Stroop effect would be present in these cases. In the actual
experiments, however, it was also possible to measure several additional
variables: Initial Rotation Direction (IDR), Rotation Direction at Peak Velocity
(RDPV), Movement Time (MT), and Reliability of constant error (R). In this
chapter, the discussion presents a possible explanation of the results for both the
original hypotheses and the additional variables. Finally, additional issues,
limitations of this study and possible future studies are mentioned.
4.1. Overview of Results
The main results of this study are summarised in Table 4.1 to assist the
discussion. This table indicates the main phenomenon under investigation, the
Spatial Stroop effect, as well as other phenomena such as Front-back reversal
and rotation effects. The table shows evidence for the Spatial Stroop effect in
Experiment One, but only for RT, and IRD, and only in the Word condition.
There is much weaker evidence for this effect in Experiment Two. In the two
experiments, there was evidence for Front-Back reversals in several variables.
The discussion will put forward an account of the overall pattern of results, and
will address the specific hypotheses where appropriate. As a tool to help explain
the results, a model of processing involved in the experimental tasks is presented
- 58 -
next. This is shown in Figure 4.1. The model contains a simplified view of the
major processes, arranged on a timeline starting before the stimulus, and ending
when the movement has been completed. The model applies to both experiments.
Table 4.1 Effects for each variable. (S denotes Spatial Stoop Effect, FBR denotes Front-Back Reversal, S? denotes possibility of Spatial Stroop Effect; X denotes no effect. L & R indicates Left and Right stimulus location only. Condition RT IRD RDPV MT CE R EXPERIMENT 1 LOCATION FBR X X X X FBR
WORD
S
(L & R) S
(ALL) X X X X EXPERIMENT 2
LOCATION FBR X X X
S? (L & R) FBR
WORD X X X X X X
4.2. Auditory Information Processing
Figure 4.1 shows a framework of the various processes that are assumed to occur
in the task required in these experiments. Numbers indicate a time scale (ms),
with a word stimulus presented at 0 ms. Before a stimulus is presented, the
subject already has a response ‘set’, that is, the instruction for that condition is to
use one dimension of the stimulus (use the relevant) and ignore the other
(irrelevant) dimension, for example, use location, ignore meaning. Two types of
information processing are started when the stimulus reaches a subject’s ears.
Auditory localisation processing indicates recognition of the stimulus location
without any reference to the meaning of the stimulus word. This processing
would commence first as it depends only on the presence of a sound and not its
specific content. Spatial semantic processing would proceed slightly later,
- 59 -
because it must decode the meaning of the word stimulus. This requires both a
longer time to sample the content of the sound, and a process to match the sound
with its memory representation. The output of these processes would then
influence the response selection stage, which would determine both the initial
movement and the final position. In all cases and both Experiments, the stimulus
lasted 500ms. In many cases, subjects had RTs less than 500ms.
Figure 4.1. Information processing model for orienting movement. (Time value is indicative only)
Time (ms)
0 1000 2000
Approximate Range for start of movement
Spatial Semantic Processing
Auditory Localisation Processing
Stimulus
Orienting Movement
RT
Instruction Response Selection
Approximate Range for end of movement
IRD MT RDPV CE
- 60 -
This suggests that the auditory localisation and/or the semantic spatial processing
can be completed in much less than 500ms, because the response selection stage
would also require some time. However, because the orienting movement can
last up to a second or more, there is still time for these processes to continue and,
possibly, modify the ongoing movement. For example, an initial movement
could be in the wrong direction and the direction must be changed, or the final
position might be decided only during the movement. The boxes in Figure 4.1
represent these processes and are therefore shown as having an unknown and
variable duration. The end result of these hypothesized processes is the execution
of an orienting movement.
The time difference between auditory localisation processing and spatial
semantic processing would make it probable that auditory localisation could
influence the initial response selection for the orienting movement, while spatial
semantic processing would influence it later on. This is in contrast to the study of
Palef and Nickerson (1978), which indicated that word processing was
undertaken faster than location processing. In the current study however, the
result showed that location processing occurred faster than word processing.
In fact, when only the Left and Right locations of the Palef and Nickerson study
are examined, their results are not so different from the current study’s outcome.
The fact that their RTs are at least 140ms slower, on average, than those for the
comparable conditions in the current study’s Experiment One, however, lends
weight to the argument that their task was more complex and required a more
cognitively demanding form of response selection. Therefore, the apparent
differences between the two studies can be explained if, in the current study, the
- 61 -
very direct and rapid form of response selection allowed the participant to begin
moving in response to the sound location before completing the processing of the
semantic content of the stimulus.
4.3. Limitations of this Study and Future Research
The present study used a whole body orienting movement as the method of
indicating the localisation judgment. However, there was no measure of gaze
position. It is possible that the head orientation and gaze position were not
completely aligned, and that slightly different outcomes would have been
obtained for actual line of sight. In addition, subjects did not receive any visual
feedback about their head orientation. Lewald, Dorrscheidt, & Ehrenstein (2000)
indicated sound localisation was more accurate when a visual indication of the
actual midline of the head was given to subjects, using a reference produced by a
laser pointer during the orienting movement. The judgment of orienting
movements may be more accurate if visual feedback for the judgment is
presented. In addition, nodding was a special case and not a natural response for
the front location. Nodding may take more time than the other movements and be
initiated more slowly.
In order to demonstrate whether spatial semantic information affected the
orienting movement in the Word condition, future experiments could use slightly
more rotated speaker positions. As described in the discussion of Experiment
Two, the relevant and the irrelevant condition (i.e. Non-rotated and Rotated
condition) may have been too small for interference from the spatial Stroop
effect. Therefore the rotated speakers could be positioned at 45 degrees, that is,
for example, in between the Left and Front speakers. When the “Left” stimulus is
- 62 -
presented from the speaker in the both Location and Word condition, the future
study hypothesises that the orienting movement will be affected by spatial
semantic information.
4.4. Conclusions
The present study found that interference caused by the spatial Stroop effect was
evident in the RT and IRD for Left and Right locations in Location condition of
Experiment One. However, this was not true for the Front and Back locations. It
was not found in later variables, and not at all in Experiment Two. These
findings allow three conclusions:
1. Spatial Stroop effect is asymmetric. The sound’s actual location can affect the
use of semantic cues, but not the reverse.
2. The spatial Stroop effect is confined to the planning and early execution of the
movement, as it is seen in only RT and initial movement direction. This provides
new information about the time course of the spatial Stroop effect.
3. These spatial Stroop effect will not occur if the location indicated by the
semantic cue and the actual location cues differ by up to twenty degrees, and are
therefore not clearly incongruent. This is true even though this rotation can
clearly be detected, evidenced by the altered final positions in Experiment Two’s
Location condition.
Another expected result was, however, confirmed, i.e. front-back reversals
occurred very frequently. More than half of the trials in the back condition were
judged coming from the front. No obvious explanation could be found for the
fact that in this study, unlike others, more back-front than front-back errors were
made.
- 63 -
References
Abel, S. M., Figueiredo, J. C., Consoli, A., Birt, C. M., & Papsin, B. C. (2002). The effect of blindness on horizontal plane sound source identification. International Journal of Audiology, 41(5), 285-292.
Abel, S. M., & Paik, J. E. S. (2004). The benefit of practice for sound localization without sight. Applied Acoustics, 65(3), 229-241.
Abernethy, B. (1991). Visual search strategies and decision-making in sports. Journal of Sports Psychology, 22, 189-210.
Begault, D. R. (1994). 3-D sound for virtual reality and multimedia. London: Academic Press Limietd.
Begault, D. R., & Wenzel, E. M. (1993). Headphone localization of speech. Human Factors, 35(2), 361-376.
Best, V., Carlile, S., Jin, C., & van Schaik, A. (2005). The role of high frequencies in speech localization. The Journal of The Acoustical Society of America, 118(1), 353-363.
Blauert, J. (1983). Spatial hearing. The Psychophysics of Human Sound Localization. Cambridge, MA: MIT Press.
Burke, K. A., Letsos, A., & Butler, R. A. (1994). Asymmetric performances in binaural localization of sound in space. Neuropsychologia, 32(11), 1409-1417.
Daus, A. T., Wilson, J., & Freeman, W. M. (1989). Predicting success in football. Journal of Sports Medicine and Physical Fitness, 29(2), 209-212.
Durlach, N. I., & Colburn, H. S. (1978). Binaural Phenomena. New York: Academic Press.
Gardner, J. D., & Sherman, A. (1995). Vision requirements in sport. Oxford: Butterworth-Heinemann.
Geierlich, H. W. (1992). The application of Binaural Technology. Applied Acoustics, 36, pp219-243.
Genuit, K. (1984). A model for the description of outer-ear transmission characteristics. Ph.D. Dissertation(Rhenish Westphalian Technical University).
Gilkey, R. H., & Anderson, T. R. (1995). The accuracy of absolute localization judgments for speech stimuli. Journal of Vestibular Research: Equilibrium & Orientation, 5(6), 487-497.
Helsen, W. F., & Sherman, A. (1999). A multidimentional apporch to skilled perception and performance in sports. Applied Cognitive Psychology, 13, 1-27.
Henning, G. B. (1980). Some obsernations on the lateralization of complex waveforms. The journal of the Acoustical Society of America., 68((2)), pp446-454.
Klatzky, R. L., Lippa, Y., Loomis, J. M., & Golledge, R. G. (2002). Learning directions of objects specified by vision, spatial audition, or auditory spatial language. Journal of Experimental Psychology. Learning, Memory, and Cognition, 9(6), 364-367.
Klatzky, R. L., Lippa, Y., Loomis, J. M., & Golledge, R. G. (2003). Encoding, learning, and spatial updating of multiple object locations specified by 3-D sound, spatial language, and vision. Experimental Brain Research, 149(1), 48-61.
- 64 -
Klump, R. B., & Eady, H. R. (1956). Some measurements of interaural time difference threshold. Journal of the Acoustical Society of America, 28, pp859-860.
Knudsen, E. I., Hasbroucq, T., & Osman, A. (1982). Visual instruction of the neural map of auditory space in the developing Optic tectum. Science, 253, pp85-87.
Kornblum, S., Hasbroucq, T., & Osman, A. (1990). Dimensional overlap: cognitive basis for stimulus-response compatibility--a model and taxonomy. Clinical Psychology Review, 97(2), 253-270.
Lewald, J., Dorrscheidt, G. J., & Ehrenstein, W. H. (2000). Sound localization with eccentric head position. Behavioural Brain Research, 108(2), 105-125.
Loomis, J. M., Lippa, Y., Golledge, R. G., & Klatzky, R. L. (2002). Spatial updating of locations specified by 3-d sound and spatial language. Journal of Experimental Psychology. Learning, Memory, and Cognition, 28(2), 335-345.
Lu, C. H., & Proctor, R. W. (1995). The influence of irrelevant location information on performance: A review of the Simon and spatial Stroop effects. Psychonomic Bulletin & Review, Vol 2(2), pp174-207.
McMorris, T. (1997). Performance of soccer players on tests of field dependence/independence and soccer-specific decision-making tests. Perceptual And Motor Skills, 85(2), 467-476.
McNamara, T. P. (2005). Semantic priming: perspectives from memory and word recognition New York: Psychology Press
Middlebrooks, J. C. (1992). Narrow-band sound localization related to external ear acoustics. Journal of The Acoustical Society of America, 92(5), 2607-2624.
Middlebrooks, J. C., & Green, D. M. (1991). Sound localization by human listeners. Annual Review of Psychology, 42, 135-159.
Middlebrooks, J. C., Makous, J. C., & Green, D. M. (1989). Directional sensitivity of sound-pressure levels in the human ear canal. Journal of the Acoustical Society of America 86(1), 89-108.
Muller, B. S., & Bovet, P. (2002). Performance and reaction times in monaural localization of first names in the horizontal plane. Brain And Language, Vol. 82(1), pp. 1-9.
Muller, H. M., & Kutas, M. (1996). What's in a name? Electrophysiological differences between spoken nouns, proper names and one's own name. Neuroreport, 8(1), 221-225.
Nakagawa, A. (1985). Present status and perspective of the study on decision making in ball games. Japanese Journal of Physical Education JAPON, 30(2), pp105-115.
Neely, J.H., (1991). Semantic priming effects in visual word recognition: A selective review of current findings and theories. In D. Besner & G. W. Humphreys (Eds.), Basic processes in reading: Visual word recognition (pp.264-336). Hillsdale, NJ: Erlbaum.
Nougier, V., & Rossi, B. (1999). The development of expertise in the orienting of attention. International Journal of Sport Psychology, Vol 30(2), pp. 246-260.
- 65 -
Oldfield, S. R., & Parker, S. P. (1984). Acuity of sound localisation: a topography of auditory space. I. Normal hearing conditions. Perception, 13(5), 581-600.
Oldfield, S. R., & Parker, S. P. (1986). Acuity of sound localisation: a topography of auditory space. III. Monaural hearing conditions. Perception, 15(1), 67-81.
Palef, S. R., & Nickerson, R. B. (1978). representing auditoy space. Perception & Psychophysics, vol.23(5), 445-450.
Paulus, E. (2003). [Sound localization cues of binaural hearing]. Laryngorhinootologie, 82(4), 240-248.
Phinney, J. S., & Nummedal, S. G. (1979). Effects of left-right orientation and position reversals on spatial perspective taking in young children. Perceptual And Motor Skills, 48(1), 223-227.
Seymour, P. H. (1974). Stroop interference with response, comparison, and encoding stages in a sentence-picture comparison task. Memory & Cognition, Vol. 2((1-A), pp. 19-26.
Shor, R. E. (1970). The processing of conceptual information on spatial directions from pictorial and linguistic symbols. Acta Psychologica, Amsterdam, Vol. 32(4), pp. 346-365.
Simon, J. R. (1990). The effects of an irrelevant directional cue on human information processing. (Vol. 65). Oxford, England: North-Holland.
Simon, J. R., Craft, J. L., & Webster, J. B. (1973). Reactions toward the stimulus source: analysis of correct responses and errors over a five-day period. Journal of Experimental Psychology, 101(1), 175-178.
Simon, J. R., & Small, A. M., Jr. (1969). Processing auditory information: interference from an irrelevant cue. Journal of Applied Psychology, 53(5), 433-435.
Starkes, J. L. (1987). Skill in field hockey: The nature of the cognitive advantage. Journal of Sport Psychology, Vol 9(2), pp. 146-160.
Stroop, J. R. (1992). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, Vol 121(1), pp. 15-23.
White, B. W. (1969). Interference in identifying attributes and attribute names. Perception & Psychophysics, 6(3), pp. 166-168.
Wightman, F. L., & Kistler, D. J. (1989a). Headphone simulation of free-field listening. I: Stimulus synthesis. Journal of the Acoustical Society of America, 85(2), 858-867.
Wightman, F. L., & Kistler, D. J. (1989b). Headphone simulation of free-field listening. II: Psychophysical validation. Journal of the Acoustical Society of America, 85(2), 868-878.
Williams, A. M., Davids, K., Burwitz, L., & Williams, J. G. (1994). Visual search strategies in experienced and inexperienced soccer players. Research Quarterly for Exercise and Sport, 65(2), 127-135.
Williams, A. M., & Grant, A. (1999). Training perception skill in sport. International Journal of Sport Psychology, 30, 194-220.
Zwislocki, J. J., & Feldman, R. S. (1956). Just noticeable differences for intensity and their relation to loudness. Journal of the Acoustical Society of America, 28, pp. 860-864.
- 66 -
Appendices
- 67 -
Appendix A: Informed Consent Form and Participant
Information Packages
- 68 -
Participant Information Sheet
“Auditory Localisation: Contributions of Sound Localisation and Semantic Spatial Cues”
Norikazu Yao
Description The purpose of this project is to investigate the effect on auditory localisation of spatial and non-spatial semantic information. The research team requests your assistance in undertaking a experiment about auditory localisation task. Your participation will involve a short screening test for hearing to make sure you have no subtle hearing impairment. Expected benefits It is expected that this project will not benefit you directly. However, it will improve our understanding of human auditory information processing and spatial orientation. Risks There are no risks associated with your participation in this project. Confidentiality All data you provide will be anonymous and will be treated confidentially. Your data will be analysed using a code and you will not be identified individually. Voluntary participation Your participation in this project is voluntary. If you do agree to participate, you can withdraw from participation at any time during the project without comment or penalty. Your decision to participate will in no way impact upon your current or future relationship with QUT. Questions / further information Please contact the researchers if you require further information about the project, or to have any questions answered. Concerns / complaints Please contact the Research Ethics Officer on 3864 2340 or [email protected] if you have any concerns or complaints about the ethical conduct of the project.
- 69 -
Participant Information Sheet
“Auditory Localisation: Contributions of Sound Localisation and
Semantic Spatial Cues”
Norikazu Yao
Statement of consent By signing below, you are indicating that you: • have read and understood the information sheet about this project; • have had any questions answered to your satisfaction; • understand that if you have any additional questions you can contact the
research team; • understand that you are free to withdraw at any time, without comment or
penalty; • understand that you can contact the research team if you have any questions
about the project, or the Research Ethics Officer on 3864 2340 or [email protected] if you have concerns about the ethical conduct of the project;
• agree to participate in the project. Name
Signature
Date / /
- 70 -
Appendix B: Statistical Analyses (6 stimuli × 4 locations) in Experiment One
- 71 -
Table 1. ANOVA table for RT in the Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 91706261 1 91706261 378.2202 0.000000 Error 2182211 9 242468 Stimulus 34028 5 6806 0.4490 0.811775 Error 682151 45 15159 Location 6028682 3 2009561 28.0943 0.000000 Error 1931289 27 71529 Stim v Loc 170560 15 11371 0.7318 0.748822 Error 2097652 135 15538
Table 2 Initial Rotation Direction for each stimulus in Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 53.49963 1 53.49963 899.2449 0.000000 Error 0.53545 9 0.05949 Stimulus 0.08875 5 0.01775 0.5910 0.706827 Error 1.35157 45 0.03003 Location 30.36618 3 10.12206 102.5998 0.000000 Error 2.66370 27 0.09866 Stim v Loc 0.26739 15 0.01783 0.6044 0.867373 Error 3.98184 135 0.02950
Table 3 Rotation Direction at Peak Velocity for each stimulus in Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 53.49963 1 53.49963 899.2449 0.000000 Error 0.53545 9 0.05949 Stimulus 0.08875 5 0.01775 0.5910 0.706827 Error 1.35157 45 0.03003 Location 30.36618 3 10.12206 102.5998 0.000000 Error 2.66370 27 0.09866 Stim v Loc 0.26739 15 0.01783 0.6044 0.867373 Error 3.98184 135 0.02950
- 72 -
Table 4 Movement Time for each stimulus in Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 275828531 1 275828531 137.6831 0.000001 Error 18030228 9 2003359 Stimulus 118523 5 23705 0.6518 0.661592 Error 1636503 45 36367 Location 1795355 3 598452 2.3822 0.091511 Error 6782959 27 251221 Stim v Loc 319190 15 21279 0.6490 0.829312 Error 4426094 135 32786
Table 5 Constant Error for each stimulus in Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 4280.996 1 4280.996 7.757701 0.021217 Error 4966.544 9 551.838 Stimulus 52.766 5 10.553 0.615614 0.688452 Error 771.413 45 17.143 Location 834.265 3 278.088 1.793036 0.172284 Error 4187.524 27 155.093 Stim v Loc 534.700 15 35.647 1.497460 0.114384 Error 3213.643 135 23.805
Table 6 Reliability for each stimulus in Location condition (6 Stimuli × 4 Location)
SS
Degree of
Freedom MS F p Intercept 197.2835 1 197.2835 1673.207 0.000000 Error 1.0612 9 0.1179 Stimulus 0.0438 5 0.0088 0.378 0.860851 Error 1.0419 45 0.0232 Location 2.6327 3 0.8776 13.654 0.000013 Error 1.7353 27 0.0643 Stim v Loc 0.2151 15 0.0143 0.726 0.755215 Error 2.6682 135 0.0198
- 73 -
Appendix C: Means and Standard Deviation Tables in Experiment One
- 74 -
Table 1. Reaction time for each stimulus in Location condition STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 434.74 431.79 831.53 743.69 431.44 418.50 817.88 769.18SD 30.42 30.69 75.87 92.63 27.33 28.69 87.40 92.26 STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 490.49 474.80 821.62 781.88 439.33 506.36 794.64 721.87SD 49.38 43.56 71.34 69.50 33.96 55.52 51.83 42.08 STIMULUS "yes" Tone LOCATION Left Right Front Back Left Right Front BackMEAN 443.92 502.21 761.06 725.40 462.31 489.03 744.50 797.49SD 24.47 59.94 88.97 45.62 26.94 37.80 42.44 39.70
Table 2 Reaction Time for each stimulus in Location and Word condition CONDITION Location Word STIMULUS "Left" "Right" "Front" "Back" "Left" "Right" "Front" "Back"MEAN 610.43 609.25 642.20 615.55 469.18 487.18 695.05 505.69SD 73.48 92.48 70.38 60.01 57.28 72.17 85.93 77.36
Table 3. Reaction Time for each location in Location and Word condition CONDITION Location Word LOCATION Left Right Front Back Left Right Front BackMEAN 449.00 457.86 816.42 754.15 549.75 523.98 548.81 534.55SD 57.50 70.52 123.53 126.19 63.89 79.17 64.07 68.64
Table 4. Reaction Time for congruent and incongruent dimension CONDITION Location Word STIMULUS "Left" "Right" "Left" "Right" LOCATION Left Right Left Right Left Right Left RightMEAN 434.74 431.79 431.44 418.50 442.30 486.51 538.26 430.77SD 30.42 30.69 27.33 28.69 30.99 43.20 45.96 35.27
- 75 -
Table 5. Initial Rotation Direction in Location and Word conditions CONDITION Location STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 0.00 0.98 0.59 0.42 0.02 0.93 0.67 0.46SD 0.00 0.02 0.06 0.10 0.02 0.05 0.10 0.10 CONDITION Location STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 0.06 0.94 0.70 0.41 0.00 0.94 0.67 0.54SD 0.03 0.03 0.08 0.09 0.00 0.04 0.09 0.10 CONDITION Word STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 0.00 0.12 0.04 0.04 0.78 1.00 0.98 0.94SD 0.00 0.08 0.03 0.03 0.08 0.00 0.02 0.04 CONDITION Word STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 0.54 0.64 0.56 0.60 0.46 0.70 0.66 0.50SD 0.10 0.08 0.11 0.13 0.08 0.10 0.11 0.11
Table 6. Rotation Direction at Peak Velocity in Location and Word condition CONDITION Location STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 0.00 1.00 0.63 0.42 0.00 0.98 0.51 0.32SD 0.00 0.00 0.09 0.07 0.00 0.03 0.09 0.10 CONDITION Location STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 0.00 0.98 0.60 0.27 0.00 0.98 0.49 0.42SD 0.00 0.02 0.08 0.05 0.00 0.02 0.08 0.08 CONDITION Word STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 0.00 0.00 0.00 0.00 0.98 1.00 1.00 1.00SD 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 CONDITION Word STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 0.52 0.58 0.54 0.50 0.50 0.62 0.64 0.44SD 0.09 0.07 0.10 0.07 0.11 0.12 0.11 0.10
- 76 -
Table 7. Movement Time for each stimulus in the Word and Location condition CONDITION Location Word STIMULUS "Left" "Right" "Front" "Back" "Left" "Right" "Front" "Back"MEAN 1115.94 1062.08 1051.51 1080.53 935.42 944.81 1012.52 1342.64SD 215.24 178.32 178.01 210.95 111.25 76.18 165.72 135.18
Table 8. Movement Time for each location in the Location condition CONDITION Location Word LOCATION Left Right Front Back Left Right Front BackMEAN 992.23 1023.39 1082.43 1212.01 1038.43 1078.30 1050.91 1067.74SD 169.26 124.79 241.75 306.80 112.29 105.04 91.40 101.81
Table 9. Constant Error for each stimuli in Location and Word condition CONDITION Location STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN 6.45 4.60 1.28 4.85 3.21 7.09 1.20 3.80SD 3.29 2.75 0.84 2.16 2.81 2.47 1.00 2.16 STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 7.35 5.11 1.71 5.98 4.61 6.78 0.94 1.82SD 3.00 2.65 1.14 3.07 3.53 2.26 0.88 1.18
CONDITION Word STIMULUS "Left" "Right" LOCATION Left Right Front Back Left Right Front BackMEAN -0.37 -0.03 -0.26 0.02 2.60 2.14 6.35 5.82SD 0.57 0.89 1.44 1.09 0.79 1.28 0.80 0.96 STIMULUS "Front" "Back" LOCATION Left Right Front Back Left Right Front BackMEAN 0.08 0.42 0.14 0.30 -0.32 0.52 2.69 1.67SD 0.51 0.47 1.07 1.08 0.73 0.41 0.80 1.39
Table 10. Reliability for each location in Location and Word condition CONDITION Location Word LOCATION Left Right Front Back Left Right Front BackMEAN 0.99 0.96 0.92 0.73 0.98 0.99 1.00 0.97SD 0.00 0.04 0.09 0.14 0.04 0.01 0.00 0.03
- 77 -
Appendix D: Statistical Analyses (2 condition × 4 stimuli × 4 locations) in Experiment One
- 78 -
Table 1. ANOVA table of Reaction time for each stimulus in Location condition (6 stimuli: “Left”, “Right”, “Front”, “Back”, “Yes” and Tone x 4 locations: Left, Right, Front, Back) SS
Table 4. Tukey HSD test for Reaction time of each location in Location condition. Error: Within MS = 70049., df = 27.000 Location Left Right Front Back Left (449.00) 0.998839 0.000171 0.000261Right(457.86) 0.998839 0.000173 0.000311Front(816.42) 0.000171 0.000173 0.720862Back(754.15) 0.000261 0.000311 0.720862
- 79 -
Table 5. ANOVA table for Initial Rotation Direction (2 condition x 4 stimuli x 4 locations)
SS
Degree of
Freedom MS F p Intercept 89.04552 1 89.04552 636.4054 0.000000 Error 1.25928 9 0.13992 Condition 0.01648 1 0.01648 0.1032 0.755360 Error 1.43754 9 0.15973 Stimulus 8.29356 3 2.76452 29.7734 0.000000 Error 2.50701 27 0.09285 Location 12.76931 3 4.25644 68.8159 0.000000 Error 1.67002 27 0.06185 Cond v Stim 7.35946 3 2.45315 23.8014 0.000000 Error 2.78282 27 0.10307 Cond v Loc 6.05480 3 2.01827 30.3380 0.000000 Error 1.79620 27 0.06653 Stim v Loc 0.16138 9 0.01793 0.6724 0.731484 Error 2.16004 81 0.02667 Cond v Stim v Loc 0.25373 9 0.02819 1.0490 0.409311 Error 2.17687 81 0.02687
Table 6. Planned comparison table for IRD with congruent and incongruent dimension (Left and Right only)
Table 7. ANOVA table for Rotation Direction at Peak Velocity (2 condition x 4 stimuli x 4 locations)
SS
Degree of
Freedom MS F p Intercept 79.07602 1 79.07602 1011.311 0.000000 Error 0.70372 9 0.07819 Condition 0.16411 1 0.16411 1.676 0.227719 Error 0.88139 9 0.09793 Stimulus 8.70127 3 2.90042 49.557 0.000000 Error 1.58023 27 0.05853 Location 11.40221 3 3.80074 72.246 0.000000 Error 1.42043 27 0.05261 Cond v Stim 11.19891 3 3.73297 42.331 0.000000 Error 2.38099 27 0.08818 Cond v Loc 8.88887 3 2.96296 59.355 0.000000 Error 1.34782 27 0.04992 Stim v Loc 0.10104 9 0.01123 0.574 0.814471 Error 1.58372 81 0.01955 Cond v Stim v Loc 0.31580 9 0.03509 1.472 0.172437 Error 1.93078 81 0.02384
- 80 -
Table 8. ANOVA table for Movement Time (2 condition x 4 stimuli x 4 locations)
SS
Degree of
Freedom MS F p Intercept 365122012 1 365122012 281.5723 0.000000 Error 11670532 9 1296726 Condition 27874 1 27874 0.0593 0.813122 Error 4233033 9 470337 Stimulus 2229545 3 743182 10.1417 0.000120 Error 1978556 27 73280 Location 658905 3 219635 2.3773 0.091982 Error 2494448 27 92387 Cond v Stim 2303368 3 767789 11.7861 0.000041 Error 1758871 27 65143 Cond v Loc 511240 3 170413 1.4720 0.244336 Error 3125831 27 115772 Stim v Loc 155473 9 17275 0.4740 0.888050 Error 2951921 81 36443 Cond v Stim v Loc 810634 9 90070 3.2228 0.002182 Error 2263783 81 27948
Table 9. Tukey HSD test for Movement Time in each stimulus in the Word condition. Error: Within MS = 1031E2, df = 27.000 Stimulus "Left" "Right" "Front" "Back" "Left" (935.42) 0.999249 0.708181 0.000187"Right"(944.89) 0.999249 0.782286 0.000197"Front"(1012.5) 0.708181 0.782286 0.000623"Back"(1342.6) 0.000187 0.000197 0.000623
Table 10. ANOVA table for Constant Error (2 condition x 4 stimuli x 4 locations)
SS
Degree of
Freedom MS F p Intercept 2449.173 1 2449.173 12.94016 0.005773 Error 1703.422 9 189.269 Condition 633.249 1 633.249 2.98561 0.118075 Error 1908.901 9 212.100 Stimulus 182.273 3 60.758 6.86407 0.001389 Error 238.992 27 8.852 Location 115.171 3 38.390 0.59740 0.622207 Error 1735.079 27 64.262 Cond v Stim 343.137 3 114.379 7.21948 0.001042 Error 427.764 27 15.843 Cond v Loc 486.158 3 162.053 3.22872 0.038011 Error 1355.156 27 50.191 Stim v Loc 136.318 9 15.146 1.15720 0.333590 Error 1060.199 81 13.089 Cond v Stim v Loc 163.507 9 18.167 1.55901 0.141830 Error 943.911 81 11.653
- 81 -
Table 11. ANOVA table for Reliability (2 condition x 4 stimuli x 4 locations)
SS
Degree of
Freedom MS F p Intercept 284.0894 1 284.0894 4738.347 0.000000 Error 0.5396 9 0.0600 Condition 0.5937 1 0.5937 7.923 0.020221 Error 0.6744 9 0.0749 Stimulus 0.0411 3 0.0137 0.782 0.514184 Error 0.4724 27 0.0175 Location 0.9484 3 0.3161 10.964 0.000069 Error 0.7785 27 0.0288 Cond v Stim 0.0447 3 0.0149 0.741 0.536769 Error 0.5426 27 0.0201 Cond v Loc 0.7733 3 0.2578 7.408 0.000897 Error 0.9395 27 0.0348 Stim v Loc 0.0459 9 0.0051 0.375 0.943957 Error 1.1012 81 0.0136 Cond v Stim v Loc 0.1648 9 0.0183 1.316 0.241536 Error 1.1271 81 0.0139
- 82 -
Appendix E: Means and Standard Deviation Tables for Yes and Tone Stimuli in Experiment One
- 83 -
Table 1. Reaction Time for "Yes" and Tone stimuli in the Location condition STIMULUS "Yes" Tone LOCATION Left Right Front Back Left Right Front Backmean 443.92 502.21 761.06 725.40 462.31 489.03 744.50 797.49SD 24.47 59.94 88.97 45.62 26.94 37.80 42.44 39.70
Table 2. Initial Rotation Direction for "Yes" and Tone Stimuli in the Location condition STIMULUS "Yes" Tone LOCATION Left Right Front Back Left Right Front Backmean 0.00 0.96 0.75 0.42 0.04 0.96 0.66 0.30SD 0.00 0.03 0.09 0.11 0.03 0.03 0.10 0.10
Table 3. Rotated Direction Peak Velocity for "Yes" and Tone stimuli in the Location condition STIMULUS "Yes" Tone LOCATION Left Right Front back Left Right Front Backmean 0.00 1.00 0.51 0.37 0.00 1.00 0.50 0.36SD 0.00 0.00 0.08 0.08 0.00 0.00 0.10 0.13
Table 4. Movement Time for "Yes" and Tone stimuli in the Location condition STIMULUS "Yes" Tone LOCATION Left Right Front Back Left Right Front Back
Table 5. Constant Error for "Yes" and Tone Stimuli in the Location condition STIMULUS "Yes" Tone LOCATION Left Right Front Back Left Right Front Backmean 3.33 8.96 1.53 3.53 3.30 4.95 0.39 8.61SD 2.72 4.59 0.87 1.96 1.55 3.21 1.24 2.98
Table 6. Reliability for "Yes" and Tone in the Location condition STIMULUS "Yes" "Tone" LOCATION Left Right Front Back Left Right Front Backmean 0.99 0.96 1.00 0.76 0.99 0.99 0.98 0.70SD 0.00 0.03 0.00 0.09 0.00 0.01 0.02 0.08
- 84 -
Table 7. ANOVA table for Stroop score of Left and Right location spatial word stimuli and Tone stimulus SS
Left Right Front Back Left Right Front Back 1116.20 1104.26 1090.08 1527.44 1041.81 1218.26 1165.91 1585.44 112.32 92.63 90.20 107.00 100.69 115.05 99.38 112.87 862.11 894.71 886.03 1285.39 814.05 958.00 941.10 1330.10
1370.30 1313.81 1294.13 1769.49 1269.58 1478.52 1390.72 1840.78 Table 4. Constant Error for each stimulus in Non-rotated and rotated condition
Location Non-
rotated Rotated
Left Right Front Back Left Right Front Back -5.11 -2.59 -5.43 6.77 7.78 6.53 -9.46 5.67 2.91 1.85 2.88 2.45 1.71 2.02 5.97 4.89
Table 5. Reliability for each stimulus in Location and Word condition
Location Word Stim-Loc Stim-Loc
Left Right Front Back Left Right Front Back 0.99 0.99 0.95 0.88 1.00 1.00 1.00 1.00 0.00 0.01 0.03 0.06 0.00 0.00 0.00 0.00 0.98 0.97 0.88 0.74 1.00 0.99 0.99 0.99 1.00 1.00 1.02 1.02 1.00 1.00 1.00 1.00
- 89 -
Appendix G: Statistical Analyses (2 condition × 2 rotation × 4 stimulus-location) in Experiment Two
- 90 -
Table 1 ANOVA table for Reaction Time (2 condition x 2 rotation x 4 stimulus-location)
SS
Degree of
Freedom MS F p Intercept 46999413 1 46999413 139.3870 0.000001 Error 3034678 9 337186 Condition 842832 1 842832 16.7630 0.002700 Error 452515 9 50279 Stimulus 7542 1 7542 0.9204 0.362436 Error 73751 9 8195 Location 1489279 3 496426 26.4888 0.000000 Error 506007 27 18741 Cond v Stim 17877 1 17877 4.5506 0.061699 Error 35356 9 3928 Cond v Loc 193317 3 64439 7.1977 0.001061 Error 241725 27 8953 Stim v Loc 10826 3 3609 0.4454 0.722557 Error 218784 27 8103 Cond v Stim v Loc 37802 3 12601 1.8536 0.161339 Error 183547 27 6798
Table 2. Planned comparison table for RT of Left Right and Front Back in the Location condition
Table 10. ANOVA table for Reliability (2 condition x 2 rotation x 4 stimulus-location)
SS
Degree of
Freedom MS F p Intercept 152.0003 1 152.0003 19825.37 0.000000 Error 0.0690 9 0.0077 Condition 0.0829 1 0.0829 9.51 0.013060 Error 0.0785 9 0.0087 Stimulus 0.0011 1 0.0011 0.39 0.549549 Error 0.0264 9 0.0029 Location 0.0764 3 0.0255 4.91 0.007495 Error 0.1400 27 0.0052 Cond v Stim 0.0025 1 0.0025 0.86 0.376963 Error 0.0256 9 0.0028 Cond v Loc 0.0727 3 0.0242 4.56 0.010369 Error 0.1435 27 0.0053 Stim v Loc 0.0044 3 0.0015 0.76 0.528601 Error 0.0522 27 0.0019 Cond v Stim v Loc 0.0047 3 0.0016 0.81 0.498462 Error 0.0516 27 0.0019
Table 11. Fisher LSD test for Reliability in each location in the Location condition. Error: Within MS = .00531, df = 27.000 Location "Left" "Right" "Front" "Back" Left (0.98861) 0.896740 0.124606 0.000078Right(0.98559) 0.896740 0.157470 0.000111Front(0.95207) 0.124606 0.157470 0.004897Back(0.88141) 0.000078 0.000111 0.004897