1 Single-Case Research Designs PS1006 Lecture 6 Sam Cromie.

1

Single-Case Research Designs

PS1006 Lecture 6

Sam Cromie

2

Group Vs Single Case Designs

Group designs:• Average performance

of a group• Comparing average

performance between groups

• Group variability • Statistical significance

Single case designs:• Actual performance of

an individual• Comparing individual

performance in different conditions

• Individual variability• Clinical significance

3

Examples of SC Resesarch• Ebbinghaus (1885) - participant and experimenter -

first systematic evidence of forgetting over time • Freud’s psychoanalytical case studies• Behaviour Analysis – SC experiments with

pigeons, monkeys, humans• Psychophysics, study of expert performance e.g.,

chess players & musicians, • Oliver Sacks - ‘The Man Who mistook His Wife for

a Hat’

4

Single case studies - characteristics

• Intensive description and analysis of single individual.

• Data obtained through: naturalistic observation, interviews, psychological tests, experimental measurement

• May describe the application and results of a particular treatment.

5

Case Studies (CS) vs Single Case Experiments (SCE)

• Exploratory• Qualitative• May generate

hypotheses for experimental research

• Experimental manipulation

• IVs and DVs• Operational definition• Measurement• Hypotheses• 6+ participants

6

Advantages of Case studies

• Provide new ideas and hypotheses – Open the way for discoveries based on

other methodologies – Provides opportunity to develop new clinical

techniques

• Try out new clinical techniques the utility of which may only become apparent in specific cases

7

• Provides chance to study rare phenomena – Infrequent occurrences can only be examined

through intensive study e.g., • Feral children - ‘The wild boy of Aveyron’ - Victor -

Lived alone in woods from ages of 5-11/12• ‘The forbidden experiment’ - Genie

– Such cases do not offer definitive answers rather ‘obliges us to reflect on how to live with these unsolved questions’ - (Shattuck, 1994, p182)

Advantages

8

• Can support or challenge scientific theories– Falsificationism

• Genie was found at 13 never having learned to communicate due to lack of human interaction.

• Lenneberg proposed critical period of language development = 2-puberty

• Lenneberg theory could be tested by determining whether Genie could now acquire language.

• Genie showed some language development but was never completely normal

• Lenneberg’s theory at the very least should be modified after the evidence ‘provided’ by Genie

Advantages

9

• Can support or challenge scientific theories– Atkinson and Shiffrin’s multi-storey theory of

memory gets considerable support from patients who show specific breakdowns in one part of the memory system.

– H.M. could have conversation and remember events for short periods of time but could not form new memories.

Advantages

10

– Individual is more than can be represented by the collection of average values on various dimensions.

– Has the ability to reveal various nuances and subtleties of behaviour that a group approach may miss.

Advantages

11

Difficulties of Case studies• Difficulty of drawing cause-effect

conclusions – Illnesses can subside spontaneously – Other aspects of the patient-therapist

relationship may have an impact– Genie - a Doctor’s examination at 14mnths lead

to the comment that she was possibly retarded - there is no way of concluding that Genie’s disposition was a product of the poor environment which she inhabited.

12

• Biases– In interpretation– Data collection. Archival records or information based

on self-reports are particularly vulnerable.

• Lack of generalisation – Difficult to generalise from case to case. – Except where it is assumed that the underlying

physiological/behavioural systems are shared e.g. psychophysics assumes that for example visual systems are based on a shared physiological makeup

Difficulties of Case studies

13

• Public often considers personal testimony as measure of efficacy – In 1980’s Laetrile, made from apricot pits,

supposed to be beneficial to the treatment of cancer.

• By using Laetrile instead of traditional therapies, many patients may have postponed valid courses of treatment and thus contributed to the spread of their cancer.

Difficulties of Case studies

14

Single-Case Experimental Designs• Core tool of Behaviour Analysis (Skinner) • Manipulating single IVs and measuring behavioural

change in one individual• Baseline compared with intervention phases• Assumes the behavioural principles are universal,

across individuals and organisms• Graphical depiction of results • Impact of results visually rather then statistically

determined – “if the difference is not obvious it is not significant”

15

Basic SC experimental designs

• A= Baseline; B= Intervention 1; C= Intervention 2• AB – weak design• ABA – reversal procedures adds to predictive power, but

ends up with baseline• ABAB – even more predictive power and ends up with

intervention• ABC, ABAC, ABCD, etc.• Problems:

– ethics of withdrawal, – intervention may have non-reversible effects – collateral

reinforcement, verbal behaviour

16

AB designJack

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

No. of Days

Inap

pro

pri

ate

Vo

caliz

atio

ns/

Hit

tin

g

Series1

BaselineToken Economy/ DRO.

A B

17

ABAB Design Experiment

18

An ABC design experiment

Junior Infants

0

20

40

60

80

100

120

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58

No of Sessions

Off

Tas

k P

erce

nt

Series1

Baseline Training Token Economy

19

Multiple baseline designs

• An independent variable is sequentially applied to at least two dependent variables

• Multiple baseline – Across behaviours– Across settings– Across participants

• Avoid problems of reversals while still demonstrating the impact of the independent variable

20

Multiple baseline design

21

Changing criterion designs

• Progressively change the level of the target behaviour required for reinforcement

• Track changes relative to the criterion

22

Changing Criterion Design

23

Readings

Concise Overview:

• Leslie & O’Reilly Behaviour Analysis: foundations and applications to psychology 1999. Chapter 8

More detail:

• Cooper,J; Heron,T; Heward,B; Applied Behaviour Analysis 2007 2nd Edition