Is startle a lateralised response in early infancy?

This article was downloaded by: [marco dondi]On: 15 August 2012, At: 00:08Publisher: Psychology PressInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

Laterality: Asymmetries of Body,

Brain and CognitionPublication details, including instructions for authors

and subscription information:

http://www.tandfonline.com/loi/plat20

Is startle a lateralised response

in early infancy?Laura Franchin

a , Sergio Agnoli

a & Marco Dondi

a

a Dipartimento di Scienze Umane and Neuroscience

Centre, University of Ferrara and National Institute of

Neuroscience, Ferrara, Italy

Version of record first published: 01 Aug 2012

To cite this article: Laura Franchin, Sergio Agnoli & Marco Dondi (2012): Is startle

a lateralised response in early infancy?, Laterality: Asymmetries of Body, Brain and

Cognition, DOI:10.1080/1357650X.2012.704038

To link to this article: http://dx.doi.org/10.1080/1357650X.2012.704038

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden.

The publisher does not give any warranty express or implied or make anyrepresentation that the contents will be complete or accurate or up todate. The accuracy of any instructions, formulae, and drug doses should beindependently verified with primary sources. The publisher shall not be liablefor any loss, actions, claims, proceedings, demand, or costs or damages

whatsoever or howsoever caused arising directly or indirectly in connectionwith or arising out of the use of this material.

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

Is startle a lateralised response in early infancy?

Laura Franchin, Sergio Agnoli, and Marco Dondi

Dipartimento di Scienze Umane and Neuroscience Centre, University

of Ferrara and National Institute of Neuroscience, Ferrara, Italy

The aim of the study was to explore whether the acoustic startle response shows

signs of early lateralisation. Using non-invasive startle measurements (Automated

Infant Motor Movement Startle Seat and Facial Action Coding System), an analysis

of response latencies and intensities on the right and left body sides was performed,

investigating the presence of asymmetries on the whole-body startle and on the

facial component of the startle motor pattern in a group of 5-month-old infants.

The findings suggest that the infant whole-body startle is a lateralised response,

characterised by a right bias latency. This lateralisation could reflect an underlying

lateralised organisation of the infant startle neural circuitry. On the other hand, the

analysis of the facial component of the startle motor pattern did not reveal any

significant asymmetry. The discrepancy found in the whole-body response and in

the startle facial component will be discussed, reflecting on the limits of the adopted

methodologies. The use of a high-speed camcorder might allow future research to

analyse more in depth the startle fast face responses.

Keywords:Whole-body startle; Automated Infant Motor Movement Startle Seat;

AIMMSS; Facial Action Coding System; FACS; Baby FACS; Asymmetries;

Infancy.

The startle reflex is generally considered to be a primitive and stereotyped

defensive reflex that is elicited by a sudden and unexpected stimulus. It

Address correspondence to: Marco Dondi, Dipartimento di Scienze Umane, University of

Ferrara, via Savonarola 38, 44100 Ferrara, Italy. E-mail: [email protected]

The authors wish to thank the children and parents who participated in this research and the

DPSS (Dipartimento di Psicologia dello Sviluppo e della Socialiszazione) of the University of

Padova for their invaluable contributions, support, and cooperation in this project. Special thanks

and gratitude is also given to Professor Francesca Simion, to Dr Piero Scatturin, for his technical

support in the apparatus development, and to the late and much missed Professor Alberto

Mazzocco, who first trusted and had faith in this project. Additional thanks to Laura Carver for

thoughtful comments on an earlier draft of this paper. At present Laura Franchin works at

Dipartimento di Psicologia dello Sviluppo e della Socialiszazione, University of Padova; while

Sergio Agnoli works at Marconi Institute for Creativity, MIC, University of Bologna.

LATERALITY, 2012, iFirst, 1�16

# 2012 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

http://www.psypress.com/laterality http://dx.doi.org/10.1080/1357650X.2012.704038

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

consists of a generalised flexion motor response that extends from the face

through the trunk and the knees (as a ‘‘cascade reaction’’; Lang, 1995). The

motor component of this reflex includes eye closure (better known as

eyeblink reflex), facial grimacing, neck flexion, arm and leg abduction or

flexion (Hunt, Clarke, & Hunt, 1936; Landis & Hunt, 1939). These motor

reactions, usually elicited by an acoustic stimulation (somatosensory and

visual modalities are less utilised), are typically supposed to be bilateral and

symmetrically distributed (Brown et al., 1991; Grosse & Brown, 2003;

Hillman, Hsiao-Wecksler, & Rosengren, 2005).

In recent years neuroanatomical research examined the lateralisation of

the neural correlates of the acoustic startle reflex, elucidating its circuitry

(Buchanan, Tranel, & Adolphs, 2004; Davis, Gendelman, Tischler, &

Gendelman, 1982; Kettle, Andrewes, & Allen, 2006; Koch & Schnitzler,

1997; Yeomans & Frankland, 1995). The startle neural circuit consists of an

ipsilateral pathway: the auditory nerve fibres, which receive the acoustic

startle probe, project ipsilaterally onto cochlear root neurons, whose thick

axon collaterals project contralaterally, crossing the brainstem and termina-

ting in the Nucleus reticularis pontis caudalis (NRPC) with outputs to the

descending reticulo-spinal tract to the spinal cord (Buchanan et al., 2004;

Davis et al., 1982; Davis, Walker, & Lee, 1999; Kettle et al., 2006). On the

other hand, focusing on the study of the possible lateralisation of the motor

responses that characterised the startle pattern, only a few studies are pre-

sently available on normal side-to-side differences of the motor startle

reflex following binaural stimulation in healthy adult participants (see

Hackley & Graham, 1987; Hillman et al., 2005; Kofler, Muller, Rinnerthaler-

Weichbold, & Valls-Sole, 2008). In 2005 Hillman and colleagues, using a

platform which provides information on the movement of the centre of foot

pressure (COP) in the anterior�posterior and medial�lateral directions,

showed that the startle postural reaction in adults is characterised by an

initial anterior movement, followed by a posterior one, and by no differences

in the medial-lateral directions. Unlike Hillman et al. (2005), a study by

Kofler et al. (2008) showed that the startle reflex is not completely

symmetrical. The aim of Kofler et al.’s work was to examine the normal

side-to-side differences of the startle reflex in bilaterally recorded muscles

(Masseter, Orbicularis oculi, Sternocleidomastoid, and Biceps brachii)

following binaural stimulation in healthy right- and left-handed adults.

The authors provided evidence of a laterality in the Sternocleidomastoid and

Biceps brachii muscles, associated with the hand dominance, and a right

Orbicularis oculi bias, irrespective of the participants’ handedness. The

authors explained these results suggesting that these asymmetries subserve

the functional role of facilitating the rotation of the head towards the

dominant side via the contralateral Sternocleidomastoid reaction, and of

facilitating the defensive response inducing more flexion movement in the

2 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

dominant arm with the ipsilateral Biceps brachii reaction. Relative to the

Orbicularis oculi asymmetry, they assumed that it may be in part related to a

differential hemispheric modulation of brainstem reflexes. However, the

facial asymmetry found in the adults’ Orbicularis oculi muscle by Kofler is

not a consistent finding between studies. The right bias was reported also by

Kettle et al. (2006), Hackley and Graham (1987), and Hager & Ekman

(1985), but it was not confirmed in the researches by Bradley, Cuthbert, and

Lang (1991, 1996).

The discrepancy between the results of Kofler et al. (2008) and Hillman

et al. (2005) on the lateralisation of the whole body startle response probably

arises from the different methodologies (COP vs EMG) used to study the

whole-body response. Moreover, Kofler et al. examined only the activity of

some single muscles involved in the motor response, while Hillman et al.

considered the global motor response, where probably the contribution of

the leg muscles (adults standing on the COP) diminished the contribution

of the lateralised muscles activities of the upper limb, highlighting only a

significant anterior movement at the beginning of the startle reaction.

Despite the importance of the startle reflex in psychophysiological

research, the neurophysiological functioning of the startle motor response

in the first months of life is yet an unexplored area. This lack is surprising,

because the startle response is considered a central psychophysiological

index to study the development of affective and attentive processes in the

first months of life (see Anthony & Graham, 1983, 1985; Balaban, 1995;

Balaban, Anthony, & Graham, 1989; Quevedo, Smith, Donzella, Schunk,

& Gunnar, 2010; Richards, 1998, 2000). Moreover, startle is already a

fundamental neurological index at birth and even at the prenatal stage to

detect the correct functioning of the neural pathways involved in this reflex

and, more in general, to assess the neurological conditions of the infant or

foetus (Divon et al., 1985; Emory & Mapp, 1988; Korner, 1969; Kuhlman,

Burns, Depp, & Sabbagha, 1988). In fact, the most important neurobeha-

vioural scales in the neonatal clinical assessment (see Amiel-Tison, 1995,

2002; Brazelton, 1973; Prechtl & Beintema, 1964; Scanlon, Nelson, Grylack

& Smith, 1979) suggest assessment of the intensity, velocity, and symmetry of

the response or, more generally, detecting the deficit (hypo-startle) or the

excessive presence (hyper-startle) of the spontaneous or elicited response

(Divon et al., 1985; Emory & Mapp, 1988; Kuhlman et al., 1988).

For the first months of life there is a considerable body of evidence of

postural and other motor biases in neonatal reflexes, such as palmar grasp

reflex, automatic walking reflex, and Babinski reflex. In most infants these

reflexes show a rightward bias in strength, coordination, and frequency

(Caplan & Kinsbourne, 1976; Grattan, De Vos, Levy, & McClintock, 1992;

Peters & Petrie, 1979). In particular, Ronnqvist and co-workers (Ronnqvist,

1995; Ronnqvist & Hopkins, 1998; Ronnqvist, Hopkins, Van Emmerik, &

EARLY LATERALISATION OF STARTLE RESPONSE? 3

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

de Groot, 1998) found that the Moro reflex in newborns displayed shorter

onset latency in the right arm than the left one. To date no studies have

attempted to verify the presence of any startle bias in early infancy. Usually

researchers and clinicians assume that it is a bilateral and symmetrically

distributed motor reaction (Brown et al., 1991; Grosse & Brown, 2003;

Hillman et al., 2005). Considering the importance of deepening knowledge

of the neurophysiological functioning of the startle motor response in early

infancy in order to better investigate the normal functioning of a widely

utilised reflex in both the clinical and neuropsychological fields, the aim of

the present study was to examine the characteristics of the infant’s startle

response, verifying whether or not this response is a symmetrical reaction

during infancy. The present work investigated the infant startle lateralisation

by analysing, in particular, the asymmetries of the whole-body startle and of

the facial component of the startle motor pattern in a group of 5-month-old

infants.

The methodologies adopted in this study are based on a completely non-

invasive and non-intrusive measurement of the infants’ startle response,

allowing an analysis of the whole-body startle and of its facial component,

separately, on the two sides of the infant’s body. Using AIMMSS (Automated

Infant Motor Movement Startle Seat), a computerised instrument for non-

invasive startle measuring (Agnoli, Franchin, & Dondi, 2011), a recording of

the intensity and latency of the whole-body motor response was realised on

the right and left body sides. This new instrument consists of an infant seat

suitable for infants of about 5 months old. Since its correct functioning was

previously tested with 5-month-olds, we chose to test a sample of infants of

the same age. Using FACS (Facial Action Coding System; Ekman, Friesen,

& Hager, 2002), particularly the version readapted to the infant’s face, the

Baby FACS (Oster, 2009), an analysis of the latency and intensity of the

muscular contractions involved on the right and left facial actions was

conducted.

METHOD

Participants

Sixteen 5-month-old infants (five girls) were involved in the study (M�19.2

weeks of age, SD�0.8). All the infants were healthy and born at full term

with birth weights appropriate for their gestational ages. They were recruited

by letters sent to parents identified through birth records. A total of 28

parents were contacted, and 16 participated in the study. All were Caucasian,

with an age range between 22 and 38 years (M�31.12 years, SD�4.32).

Written informed consent was obtained from the infants’ parents.

4 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

Startle data acquisition and measurement

Whole-body startle. The Automated Infant Motor Movement Startle Seat

(AIMMSS) is a new non-invasive instrument developed by the research

group (Agnoli et al., 2011) to perform a registration of the infant whole-

body reactions to startling stimuli. In particular in this study it was

employed to detect and analyse whole-body startle asymmetries. Startle

motor responses were measured by extensimetric sensors (strain gauges)

glued onto four stainless steel cantilevers mechanically coupled to an infant

seat and electrically connected to a dedicated preamplifier. The mechanical

deformations of the four cantilevers, resulting from a whole-body startle

response, produced a voltage signal that was integrated by the amplifier to

produce an output signal proportional to the change of strain applied by the

child’s movement on the infant seat.

Time series from the left and right channels were acquired with a Labview

DAQ card and a customised Labview program (National Instruments Corp,

Austin, TX), and were stored on a laptop for off-line analysis. In the same

Labview graphic programming environment the onset latency and amplitude

of the left and right AIMMSS signals were calculated. Onset latency was

defined as the time between the probe stimulus onset and the time at which

the startle response began. Startle amplitude was defined by the maximum

voltage generated within 600 ms of the stimulus onset (see also Hillman

et al., 2005; Hillman, Rosengren, & Smith, 2004; Winslow, Parr, & Davis,

2002).

Facial component of the startle motor pattern. The Facial Action Coding

System (FACS; Ekman et al., 2002) and Baby FACS (Oster, 2009), a version

of FACS applicable to infants and neonates, were used to assess the latency

and intensity of facial muscular activity involved in the startle response,

separately on the right and left sides of the face. FACS coding consists in

decomposing a facial movement into the particular Action Units (AUs) that

produced it, either singly or in combination with other units (Ekman,

Friesen, & Simons, 1985). Specifically, micromeasurement was made for each

AU that appeared in the face as consequence of acoustic stimulation. Two

trained FACS coders (the primary coder was a certified FACS coder; the

second coder, although not certified, was trained to code the action units

described below in the text, for reliability purposes) independently watched

the videos in slow motion and, where necessary, frame by frame. As in

Ekman et al. (1985) and Hager and Ekman (1985), the two coders analysed

the onset and intensity of the actions involved in the facial component of the

startle response: Brow Lowerer (AU4), Cheek Raiser (AU6), Lid Tightner

(AU7), and Blink-Optional (AU45). Moreover, according to Baby FACS

EARLY LATERALISATION OF STARTLE RESPONSE? 5

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

guidelines, Brow Knitting (AU3) was analysed.1 The Action Units 20 (Lip

Stretcher) and 21 (Neck Tightener), identified by Ekman et al. (1985) as

actions involved in the facial component of the startle response, were not

analysed because (in line with Oster, 2009) these two Action Units are

difficult to code on an infant’s face. In fact the typical infant body shape,

with a shorter and more thickly padded neck compared with that of adults,

makes the observation of the Lip Stretcher and Neck Tightener difficult

(Oster, 2009). This issue was also confirmed in a previous work on 5-month-

old infants (see Agnoli et al., 2011). Following Ekman et al.’s (1985) scoring

procedure, any of these actions (AU3; AU4; AU6; AU7; AU45) that began in

the first 200 ms after the acoustic startle probe were scored, since this

interval includes the beginning of actions that comprise a characteristic

startle response (see Hager & Ekman, 1985). Because the film had been

exposed at 25 frames per second, specifying the number of frames would

provide data in 40-ms units.

Startle AUs onset and intensity coding was conducted separately for each

side of the face, using custom software (Adobe Premiere Pro). The survey of

the AUs characteristics was independently conducted on each side of the

face, temporarily obscuring the side not submitted to analysis (Hager

& Ekman, 1985). Video-recordings were viewed extensively in slow, frame

by frame and real-time motion to distinguish the AUs onset and intensity.

To find the onset time of each AU, coders proceeded until the movement

peaked or was clearly visible. They then moved the recording backward until

the movement stopped and noted the onset time. They confirmed or adjusted

the onset time by moving the recording forward and backward across the

estimated onset point. After each AU onset identification, the two coders

calculated the AU onset latency on both sides of the face by counting the

number of frames elapsing between the acoustic stimulation and the

muscular contraction onset.

In addition the coders recorded, separately for the two sides of the face,

when each AU reached maximum muscular contraction (apex). The coders

rated the AUs’ apex intensity from ‘‘A’’ to ‘‘E’’ level, following FACS and

Baby FACS guidelines: A level refers to a trace of the action, B to slight

evidence, C to a marked or pronounced action, D to a severe or extreme

action, and E to the maximum evidence. However, for every AU the criteria

for each intensity level in terms of the scale of evidence are described in the

FACS coding manual.

1Baby FACS (Oster, 2009) distinguishes between Corrugator supercilii action, coded as AU3

(knitting of the brow due to medial contraction), and Procerus action, coded as AU4 (knotting of

the brow due to lowering of glabella). The adult FACS (Ekman et al., 2002) does not separately

code knitting but includes it in AU4 (Oster, 2005).

6 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

The extent of intercoder agreement was evaluated through three indexes:

(a) identification of each AU, (b) location of movement beginning, and (c)

judgement of the intensity of the actions for each side of the face. Intercoder

reliability (Cohen’s Kappa) based on about 80% of all probes (N�96) was

high for the identification of AUs (k�.83) on the left and right side of the

face, and it ranged from .68 to .90 for the location of movement onset. In

accordance with Ekman et al. (1985) we calculated intercoder agreement on

intensity ratings separately for each of the AUs codified on the two sides of

the face that were shown by half or more of the participants (.57Bk�.79).

In cases of important discrepancies in codes, coders reviewed and assigned

scores by consensus. A blink was considered invalid if the probe occurred in

the midst of a naturally occurring blink.

Stimuli

The acoustic startle probe was a 50-ms burst of white noise (20 Hz � 20 kHz)

presented at an intensity of 95 dB, with instantaneous rise/fall time. It was

generated by a custom white noise generator and presented binaurally

through two speakers at about 50 cm from the infant’s left and right ear,

respectively.

Procedure

Parents were asked to bring their infants to the laboratory about an hour

before an anticipated feeding so the infants could be tested when awake, but

not hungry. Before starting the session experimenters explained the

procedure to the parents. The infant was seated on the infant seat

(AIMMSS) and the parent was seated at the side where she/he was

encouraged to interact freely with her/his child. The parents’ position was

randomised across infants. Experimenters and equipment were separated

from parents and infants by room dividers that totally isolated the dyad

during the experimental trials. During the experimental session two digital

camcorders on tripods recorded the infant’s face and whole body, respec-

tively. The first camera was used for collecting data for facial coding. The

other camera was used for reviewing offline each experimental trial and for

controlling signal artefacts possibly caused by the infant’s (or by the

parent’s) motor activity not related to the acoustic probes. The height of

the tripods and the angle of the cameras were adjusted to obtain a level,

straight-on view of the child’s face and body.

Six acoustic probes were presented with an inter-trial interval of at least

10 seconds. The startle probes were presented without notice by one of the

experimenters when the spontaneous motor activity of the infant was

reduced to a minimum, the infant’s body weight was centrally distributed on

EARLY LATERALISATION OF STARTLE RESPONSE? 7

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

the infant seat, and the head was in the centre between the two speakers. The

other experimenter sat silently behind the cameras to ensure a correct video-

recording. The stimulus presentation was marked by a red LED (with a

blinking duration of 50 ms), placed on the top of the infant seat and visible

in shot of the cameras, to facilitate the identification of the stimuli onset in

the video coding.

RESULTS

A total of 96 probe presentations (6 probes per 16 infants) were considered

for data analysis. For the whole-body startle analysis, 37 events (38.5% of the

all probes) produced analysable startle signals. In the remaining probes, even

when using a frame-by-frame video analysis, infants did not exhibit any

whole-body motor response. Infants were included in the statistical analysis

only if they exhibited at least one whole-body startle. For the analysis of the

facial component of the startle motor pattern a valid video data recording

was obtained for 90.6% (n�87) of the probes. Nine un-analysable probes

(9.4%) were excluded from analysis because there was not a straight-on shot

of infant’s face (n�5), or the probe occurred in the midst of a naturally

occurring blink (n�1) or when the Action Units considered in the present

study were already activated (n�3). Within the analysable probes (n�87),

65 events (74.7%) produced at least one analysable Action Unit activity,

while 22 events (25.3%) did not present any Action Unit linked to the startle

responses.

The dependent variables considered for statistical analyses were: whole-

body startle onset latency and amplitude derived from AIMMSS left and

right channels registration, and AUs onset latency and intensity derived from

FACS coding on the right and left sides of the face. Mean values for each

infant (n�16) were analysed. In addition, the data were also analysed using

a probe-by-probe analysis of all analysable events (n�37 for the whole-body

startle; n�65 for the startle facial component). In order to analyse infants’

startle asymmetries, data derived from left and right body sides were

compared. Finally, we analysed whether startle habituated to the acoustic

stimulation and whether the habituation affected the response lateralisation.

Statistical significance levels used a criterion of pB.05.

Whole-body startle

Comparing the mean latencies derived from the right and left AIMMSS

channels, the analysis revealed a significant difference, t(15)��2.42,

p�.029, d��0.31, 95% CI (�26.71, �1.69). The startle response latency

recorded by the right channel (M�145.47; SD�45.48) was significantly

shorter than the latency recorded on the left channel (M�159.67;

8 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

SD�44.51), as Figure 1(a) shows. Also the probe-by-probe analysis

confirmed this significant trend, t(36)��2.51, p�.017, d��0.26, 95%

CI (�32.94, �3.52). Regarding the analysis of the amplitudes derived from

the right and left channels, the response intensity recorded on the two body

sides (right: M�0.053; SD�0.049; left: M�0.044; SD�0.038) showed no

significant differences in both analyses (Figure 1b).

In order to explore the startle response habituation we split the six probes

into two blocks, including the first three probes of each infant in the first

Figure 1. Startle parameters (Panel a: onset latency; Panel b: amplitude) registered by the left and

right AIMMSS channels. (a) Mean onset latencies (n�16), expressed in milliseconds. (b) Mean

amplitudes (n�16), expressed in Volt. Error bars represent standard deviations.

EARLY LATERALISATION OF STARTLE RESPONSE? 9

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

block and the latter three (probes 4, 5, 6) in the second block. The

habituation effects on startle latency and amplitude asymmetry was explored

through a repeated-measure ANOVAwith Laterality (2 levels: right; left) and

Block (2 levels: first; second) as within-participant factors.

For the latency parameter the results showed a Laterality effect,

F(1, 13)�4.871, p�.046, hp2�.273, that confirmed the asymmetry in startle

latency already described above, with a shorter latency on the right side.

No effect of Block, F(1, 13)�1.436, p�.252, hp2�.099, and no interaction

between Laterality and Block, F(1, 13)�.508, p�.489, hp2�.038, emerged.

For the amplitude parameter the results showed only a Block effect,

F(1, 13)�9.161, p�.010, hp2�.413, that highlighted a stronger amplitude in

the first block. No effect of Laterality, F(1, 13)�1.466, p�.247, hp2�.101,

and no interaction between Laterality and Block, F(1, 13)�4.112, p�.064,

hp2�.240, emerged.

Facial component of the startle motor pattern

Mean values and standard deviations of the right and left startle AUs latency

and intensity coded using Baby FACS are reported in Table 1. Only Action

Units 7 and 45 were detected in the entire sample; AU 6 was detected in the

majority of the babies’ faces, while Action Units 3 and 4 were detected in

about half of the sample (see Table 1). The analyses performed on the AUs

latencies and intensities of the facial component for each Action Unit failed

to reveal any significant difference between the two sides of the face.

DISCUSSION

The aim of the present study was to determine whether the acoustic startle in

5-month-old infants is a symmetric or lateralised motor reflex. The findings

suggest that the whole-body startle resulting from binaural stimulation is

lateralised towards the right side. A whole-body lateralisation emerged from

the analysis of the startle asymmetries, based on the AIMMSS. The data

revealed a difference in the onset latencies between the right and left body

sides. In particular, the startle latency recorded on the right AIMMSS

channel was significantly shorter than the latency recorded on the left

AIMMSS channel. However, in the analysis of the amplitudes registered on

the two body sides, no significant difference was found. The analysis of the

facial component of the startle motor pattern, performed using the Baby

FACS coding system (Oster, 2009), also showed no significant difference in

the Action Units latencies and intensities computed on the two face sides. In

addition an habituation effect in startle amplitude (a reduction in amplitude

across probes) was found. This result is not surprising, considering that

the whole-body startle response is usually characterised by a rapid

10 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

habituation (Cassella & Davis, 1986; Hillman et al., 2005; Winslow et al.,

2002). However, the statistical analyses showed that this habituation did not

affect the lateralisation of the response, neither in the amplitude nor in the

latency parameter.

These findings are consistent with the studies on infancy that analysed the

lateralisation of other reflex motor activities. The lateralisation of the startle

latency, that is the shorter latency recorded on the right body side, suggests

that the 5-month-old infants’ startle motor response is characterised by a

right bias latency, as Ronnqvist and co-workers (Ronnqvist, 1995; Ronnqvist

& Hopkins, 1998; Ronnqvist et al., 1998) found in the abduction/extension

movements of the Moro reflex in healthy full-term newborns. However, the

lack of significant results regarding the intensity parameter of the whole-

body response is in line with the research on infants of Ronnqvist and

colleagues and with the research on adults’ startle of Hillman et al. (2005).

On the other hand, the absence of significant data in the analysis of the facial

component of the startle motor pattern in our study is not comparable with

any other infants’ research. The only data present in the literature on

eyeblink lateralisation regards a stronger right bias found in the adults’

Orbicularis oculi muscle detected with the FACS system by Hager and

Ekman (1985), and with the EMG by Kettle et al. (2006), Kofler et al.

(2008), and Hackley and Graham (1987), even if it has not been confirmed

by Bradley et al. (1991, 1996). Probably the Orbicularis oculi bias found in

the majority of studies on the lateralisation of adults’ startle motor pattern

could be not yet completely developed in 5-month-old infants. Alternatively

we could hypothesise that the lack of significant facial asymmetries could

arise from the infants’ face morphological characteristics, and in particular

from the subcutaneous fat that reduces or eliminates fine lines and wrinkles

TABLE 1

Mean (SD) latency and intensity of startle Action Units coded using Baby FACS

AU3 AU4 AU6 AU7 AU45

n Right Left n Right Left n Right Left n Right Left n Right Left

Latency 8 132 132 7 139 139 10 63 63 16 83 82 16 67 70

(47) (47) (20) (20) (22) (22) (22) (21) (14) (14)

Intensity 8 1.56 1.56 7 1.80 1.80 10 2.53 2.53 16 2.53 2.53 16 3.80 3.80

(0.42) (0.42) (0.58) (0.58) (0.84) (0.84) (0.74) (0.78) (0.54) (0.55)

Action Units (AUs) parameters derived from the Baby FACS codings on the left and right side

of the face. n indicates the number of infants who exhibited at least once the AU indicated in each

column. AUs latency is expressed in milliseconds. FACS intensity alphabetic values were

transposed in numeric values that ranged from 1 (trace of the action) to 5 (maximum evidence).

EARLY LATERALISATION OF STARTLE RESPONSE? 11

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

(Oster, 2009), causing serious difficulties in the video analysis of the different

intensities of (left vs right hemiface) muscle contractions.

Adopting a neurophysiological perspective, we suggest that the right

startle motor bias showed by 5-month-olds could reflect an underlying

lateralised organisation of the startle neural circuitry, in particular at the

spinal level. Ronnqvist (1995) suggested that the Moro reflex lateralisation

derives from some spinal asymmetries. The interpretation of Ronnqvist

(1995) would also be helpful for an understanding of the rightward startle

body bias. The startle neural circuitry is characterised by the activation of

three synapses at the level of cochlear root neurons, neurons in the NRPC,

and motoneurons in the facial motor nucleus and spinal cord (Davis, 1997;

Davis et al., 1999; Lang, Davis, & Ohman, 2000). The reticular formation

influences motor activities through its reciprocal connection with red nucleus,

substantia nigra, subthalamus, basal ganglia, motor cortex, cerebellum, and

spinal cord (Traurig, 2008). On the basis of a cross-species similarity of the

startle circuitry (Yeomans & Frankland, 1995), Davis et al. (1982) demon-

strated that the startle circuitry consists of an ipsilateral pathway from the

auditory nerve to the ventral cochlear nucleus, which crosses the brainstem

to the contralateral NRPC. The NRPC sends output to the contralateral

facial motor nucleus, which contains the motoneurons that innervate the

Orbicularis oculi muscle (Meloni & Davis, 1992). On the basis of the lack of

significant results in the analysis of the startle facial component, it seems

that this neural tract does not present asymmetries in 5-month-old infants.

Apart from the contralateral facial nucleus, the NRPC also sends output to

the spinal cord. Specifically, the Pontine reticular nucleus and the Gigan-

tocellular nucleus of the medulla provide predominantly ipsilateral and

contralateral projections to spinal cord via the medial reticulospinal and

lateral reticulospinal tracts, respectively (Traurig, 2008). Pontine reticular

nucleus facilitates motor neurons that innervate axial and limb extensor

muscles. Gigantocellular nucleus projections exert facilitatory effects on

neurons innervating limb flexors and inhibit lower motor neurons innerva-

ting axial and limb extensor muscles. Considering that the whole-body

startle consists of a neck flexion, and arm and leg abduction or flexion (Hunt

et al., 1936; Landis & Hunt, 1939), the Gigantocellular nucleus projections

might contribute to the rightward body bias found in this motor reaction,

exerting facilitatory effects on neurons innervating limb flexors. Therefore,

similar to the Moro reflex, the startle rightward body bias found at 5 months

old could be determined by the presence of some asymmetries at the spinal

level.

Nevertheless this interpretation should be considered cautiously, due to

the relatively small sample size investigated, and to the small number of

analysable responses of the whole-body startle. The choice of measuring the

whole-body startle response involves the typical issue of the low number of

12 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

analysable responses. In fact, whole-body startle response is subject to a

stronger habituation than startle blink, which is the response component

usually measured in startle research (Blumenthal et al., 2005). Indeed, both

in animals and in adult humans, the whole-body startle is characterised by

lower response rate than blink reflex (Cassella & Davis, 1986; Hillman et al.,

2005; Winslow et al., 2002). In accordence with these data, in the present

work and in a previous study (Agnoli et al., 2011), we also found a low

whole-body startle frequency in 5-month-old infants. However, the results of

the present work could represent a first insight for a deeper analysis of infant

startle asymmetries that could bring new data to the study of the startle

neurological organisation in the first months of life.

A final consideration regards the discrepancy found in the results on the

latency parameter obtained by using the AIMMSS and Baby FACS coding

system. Baby FACS coding was executed on the video-recordings performed

by a standard camcorder. The FACS methodology has already demonstrated

its reliability in detecting adults’ asymmetric facial behaviours (see Hager &

Ekman, 1985). Consequently we could hypothesise that the lack of detection

of the asymmetries in the latency parameter could be imputed to a technical

limit of the standard camcorder used to record the startle facial component,

rather than to the coding procedure used to measure it. The discrepancy in

the detection of asymmetries in the speed parameter can be explained by the

fact that the two recording instruments (AIMMSS and camcorder) are

characterised by a different sensitivity; that is, a different temporal

resolution. While the camcorder used in this study for the Action Units

onset video coding recorded at 25 frames/second*that is, one data (frame)

every 40 milliseconds*the AIMMSS recorded data with a time sampling of

5 milliseconds. Given that the full-body latency differences were on the order

of 15 ms, it is possible that a small latency difference might be obscured by

up to 40 ms of jitter in the face response latency data. Thus the discrepancy

in the results regarding the latency parameter might be influenced by the

different time sampling of the instruments adopted in this work. We propose

as a possible solution for future research to combine the use of Baby FACS

coding with the use of more sensitive instruments such as a high-speed

camcorder. In this way it should be possible to obtain a direct comparison

between AIMMSS and Baby FACS AUs latencies. Moreover, future studies

with a high-speed video could better examine whether there is greater

evidence for asymmetry in quicker or more intense facial actions. Finally, it

would be very interesting to study the startle motor reactions of infants at

younger ages in order to also clarify the development of the startle circuitry.

Manuscript received 1 February 2012

Revised manuscript received 10 June 2012

First published online 31 July 2012

EARLY LATERALISATION OF STARTLE RESPONSE? 13

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

REFERENCES

Agnoli, S., Franchin, L., & Dondi, M. (2011). Three methodologies for measuring the acoustic

startle response in early infancy. Developmental Psychobiology, 53, 323�329. doi: 10.1002/

dev.20528.

Amiel-Tison, C. (1995). Clinical assessment of the infant nervous system. In M. I. Levene & R. J.

Lilford (Eds.), Fetal and neonatal neurology and neurosurgery (2nd ed.) (pp. 83�104). London:

Churchill Livingstone.

Amiel-Tison, C. (2002). Update of the Amiel-Tison neurologic assessment for the term neonate or

at 40 weeks corrected age. Pediatric Neurology, 27, 196�212.

Anthony, B. J., & Graham, F. K. (1983). Evidence for sensory-selective set in young infants.

Science, 220(4598), 742�744. doi: 10.1126/science.6836312

Anthony, B. J., & Graham, F. K. (1985). Blink reflex modification by selective attention: Evidence

for the modulation of ‘‘automatic’’ processing. Biological Psychology, 21, 43�59. doi:10.1016/

0301-0511(85)90052-3

Balaban, M. T. (1995). Affective influences on startle in five-month-old infants: Reaction to facial

expression of emotion. Child Development, 66, 28�36. doi: 10.1111/j.1467-8624.1995.tb00853.x

Balaban, M. T., Anthony, B. J., & Graham, F. K. (1989). Prestimulation effects on blink and

cardiac reflexes of 15-month human infants. Developmental Psychobiology, 22, 115�127. DOI:

10.1002/dev.420220203

Blumenthal, T. D., Cuthbert, B. N., Filion, D. L., Hackley, S., Lipp, O. V., & Van Boxtel, A. (2005).

Committee report: Guidelines for human startle eyeblink electromyographic studies. Psycho-

physiology, 42, 1�15. doi: 10.1111/j.1469-8986.2005.00271.x

Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (1991). Startle and emotion: Lateral acoustic

probes and the bilateral blink. Psychophysiology, 28, 285�295. doi: 10.1111/j.1469-8986.1991.

tb02196.x

Bradley, M. M., Cuthbert, B. N., & Lang, P. J. (1996). Lateralised startle probes in the study of

emotion. Psychophysiology, 33, 156�161. doi: 10.1111/j.1469-8986.1996.tb02119.x

Brazelton, T. B. (1973). Neonatal behavioral assessment scale. Philadelphia: JB Lippincott.

Brown, P., Rothwell, J. C., Thompson, P. D., Britton, T. C., Day, B. L., & Marsden, C. D. (1991).

New observations on the normal auditory startle reflex in man. Brain, 114, 1891�1902.

doi: 10.1093/brain/114.4.1891

Buchanan, T. W., Tranel, D., & Adolphs, R. (2004). Anteromedial temporal lobe damage blocks

startle modulation by fear and disgust. Behavioral Neuroscience, 118, 429�437. doi: 10.1037/

0735-7044.118.2.429

Caplan, P. J., & Kinsbourne, M. (1976). Baby drops the rattle: Asymmetry of duration of grasp by

infants. Child Development, 47, 532�534. doi: 10.2307/1128815

Cassella, J. V., & Davis, M. (1986). The design and the calibration of a startle measurement system.

Psychology and Behavior, 36, 377�383. doi: 10.1016/0031-9384(86)90032-6

Davis, M. (1997). The neurophysiological basis of acoustic startle modulation: Research on fear

motivation and sensory gating. In P. J. Lang, R. F. Simons, & M. Balaban (Eds.), Attention and

orienting (pp. 69�96). Hillsdale, NJ: Lawrence Erlbaum Associates Inc.

Davis, M., Gendelman, D. S., Tischler, M. D., & Gendelman, P. M. (1982). A primary acoustic

startle circuit: Lesions and stimulation studies. The Journal of Neuroscience, 2, 791�805.

Davis, M., Walker, D. L., & Lee, Y. (1999). Neurophysiology and neuropharmacology of startle and

its affective modulation. In M. E. Dawson, A. M. Schell, & A. H. Bohmelt (Eds.), Startle

modification. Implication for neuroscience, cognitive science, and clinical science (pp. 95�113).

Cambridge, UK: Cambridge University Press.

Divon, M. Y., Platt, L. D., Cantrell, C. J., Smith, C. V., Yeh, S. Y., & Paul, R. H. (1985). Evoked

fetal response: A possible intrauterine neurological examination. American Journal of

Obstetrics and Gynecology, 153, 454�456.

14 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

Ekman, P., Friesen, W. V., & Hager, J. C. (2002). Facial Action Coding System (FACS). Salt Lake

City, UT: Research Nexus division of Network Information Research Corporation.

Ekman, P., Friesen, W. V., & Simons, R. C. (1985). Is the startle reaction an emotion? Journal of

Personality and Social Psychology, 49, 1416�1426. doi: 10.1037/0022-3514.49.5.1416

Emory, E. K., & Mapp, J. R. (1988). Effects of respiratory distress and prematurity on spontaneous

startle activity in neonates. Infant Behavior and Development, 11, 71�81. doi:10.1016/S0163-

6383(88)80017-1

Grattan, M. P., De Vos, E., Levy, J., & McClintock, M. K. (1992). Asymmetric action in the human

newborn: Sex differences in patterns of organisation. Child Development, 63, 273�289. doi:

10.1111/j.1467-8624.1992.tb01626.x

Grosse, P., & Brown, P. (2003). Acoustic startle evokes bilaterally synchronous oscillatory EMG

activity in the healthy human. Journal of Neurophysiology, 90, 1654�1661. doi: 10.1152/

jn.00125.2003

Hackley, S. A., & Graham, F. K. (1987). Effects of attending selectively to the spatial position of

reflex-eliciting and reflex-modulating stimuli. Journal of Experimental Psychology: Human

Perception and Performance, 13, 411�424. doi: 10.1037/0096-1523.13.3.411

Hager, J. C., & Ekman, P. (1985). The asymmetry of facial actions is inconsistent with models of

hemispheric specialisation. Psychophysiology, 22, 307�318. doi: 10.1111/j.1469-8986.1985.

tb01605.x

Hillman, C. H., Hsiao-Wecksler, E. T., & Rosengren, K. S. (2005). Postural and eye-blink indices of

the defensive startle reflex. International Journal of Psychophysiology, 55, 45�49. doi:10.1016/

j.ijpsycho.2004.06.002

Hillman, C. H., Rosengren, K. S., & Smith, D. P. (2004). Emotion and motivated behaviour:

Postural adjustments to affective picture viewing. Biological Psychology, 66, 51�62.

doi:10.1016/j.biopsycho.2003.07.005

Hunt, W. A., Clarke, F., & Hunt, E. (1936). Studies of startle pattern: Infants. The Journal of

Psychology, 2, 339�352.

Kettle, J. W. L., Andrewes, D. G., & Allen, N. B. (2006). Lateralisation of the startle reflex circuit in

humans: An examination with monaural probes following unilateral temporal lobe resection.

Behavioral Neuroscience, 120, 24�39. doi: 10.1037/0735-7044.120.1.24

Koch, M., & Schnitzler, H. U. (1997). The acoustic startle response in rats � circuits mediating

evocation, inhibition and potentiation. Behavioural Brain Research, 89, 35�49. doi:10.1016/

S0166-4328(97)02296-1

Kofler, M., Muller, J., Rinnerthaler-Weichbold, M., & Valls-Sole, J. (2008). Laterality of auditory

startle responses in humans. Clinical Neurophysiology, 119, 309�314. doi:10.1016/j.clinph.

2007.11.006

Korner, A. F. (1969). Neonatal startles, smiles, erections, and reflex sucks as related to state, sex,

and individuality. Child Development, 40, 1039�1053. doi: 10.2307/1127010

Kuhlman, K. A., Burns, K. A., Depp, R., & Sabbagha, R. E. (1988). Ultrasonic imaging of normal

fetal response to external vibratory acoustic stimulation. American Journal of Obstetrics and

Gynecology, 158, 47�51.

Landis, C., & Hunt, W. A. (1939). The startle pattern. New York: Farrar & Rinehart.

Lang, P. J. (1995). The emotion probe. American Psychologist, 50, 372�385.

Lang, P. J., Davis, M., & Ohman, A. (2000). Fear and anxiety: Animal models and human cognitive

psychophysiology. Journal of Affective Disorder, 61, 137�159. doi:10.1016/S0165-0327(00)00

343-8

Meloni, E. G., & Davis, M. (1992). Anatomical connections mediating the pinna component of the

acoustic startle reflex in the rat. Society for Neuroscience Abstracts, 18, 1547.

Oster, H. (2005). The repertoire of infant facial expressions: An ontogenetic perspective. In J. Nadel

& D. Muir (Eds.), Emotional development (pp. 261�292). New York: Oxford University Press.

EARLY LATERALISATION OF STARTLE RESPONSE? 15

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2

Oster, H. (2009). Baby FACS: Facial Action Coding System for Infants and Young Children.

Unpublished monograph and coding manual. New York University.

Peters, M., & Petrie, B. F. (1979). Functional asymmetries in the stepping reflex of human neonates.

Canadian Journal of Psychology, 33, 198�200. doi: 10.1037/h0081717

Prechtl, H. F. R., & Beintema, D. (1964). The neurological examination of the full term newborn

infant. London: Heinemann.

Quevedo, K., Smith, T., Donzella, B., Schunk, E., & Gunnar, M. (2010). The startle response:

Developmental effects and a paradigm for children and adults. Developmental Psychobiology,

52, 76�89. doi: 10.1002/dev.20415

Richards, J. E. (1998). Development of selective attention in young infants: Enhancement and

attenuation of startle reflex by attention. Developmental Science, 1, 45�51. doi: 10.1111/1467-

7687.00011

Richards, J. E. (2000). Development of multimodal attention in young infants: Modification of the

startle reflex by attention. Psychophysiology, 37, 65�75. doi: 10.1111/1469-8986.3710065

Ronnqvist, L. (1995). A critical examination of the Moro response in newborn infants � symmetry,

state relation, underlying mechanisms. Neuropsychologia, 33, 713�726. doi:10.1016/0028-3932

(95)00006-O

Ronnqvist, L., & Hopkins, B. (1998). Head position preference in the human newborn: A new

look. Child Development, 69, 13�23. doi: 10.1111/j.1467-8624.1998.tb06129.x

Ronnqvist, L., Hopkins, B., Van Emmerik, R., & de Groot, L. (1998). Lateral biases in head

turning and the Moro response in the human newborn: Are they both vestibular in

origin? Developmental Psychobiology, 33, 339�349. doi: 10.1002/(SICI)1098-2302(199812)33:4

B339::AID-DEV5�3.0.CO;2-R

Scanlon, J. W., Nelson, T., Grylack, I., & Smith, Y. (1979). A system of newborn physical

examination. Baltimore: University Park Press.

Traurig, H. H. (2008). The brain stem reticular formation. In M. Conn (Ed.), Neuroscience in

medicine (pp. 273�286). London: Springer.

Winslow, J. T., Parr, L. A., & Davis, M. (2002). Acoustic startle, prepulse inhibition, and fear-

potentiated startle measured in rhesus monkeys. Biological Psychiatry, 51, 859�866.

doi:10.1016/S0006-3223(02)01345-8

Yeomans, J. S., & Frankland, P. W. (1995). The acoustic startle reflex: Neurons and connections.

Brain Research Reviews, 21, 301�314. doi:10.1016/0165-0173(96)00004-5

16 FRANCHIN, AGNOLI, DONDI

Do

wn

load

ed b

y [

mar

co d

on

di]

at

00

:08

15

Au

gu

st 2

01

2