A Stimulus Control Analysis of the Misinformation Effect/67531/metadc... · misinformation effect...

APPROVED: Jesus Rosales- Ruiz, Major Professor Jonathan Pinkston, Committee Member Manish Vaidya, Committee Member Richard Smith, Chair of the Department of

Behavior Analysis Thomas Evenson, Dean of the College of

Public Affairs and Community Service

Mark Wardell, Dean of the Toulouse Graduate School

A STIMULUS CONTROL ANALYSIS OF THE MISINFORMATION EFFECT

Kelly M. Tait, B.S.

Thesis Prepared for the Degree of

MASTER OF SCIENCE

UNIVERSITY OF NORTH TEXAS

August 2014

Tait, Kelly M. A Stimulus Control Analysis of the Misinformation Effect. Master of

Science (Behavior Analysis), August 2014, 43 pp., 8 tables, 9 figures, references, 42 titles.

This paper explores research on the misinformation effect and hypothesizes a new

explanation for the occurrence misinformation effect. Current psychological theories states the

misinformation effect occurs when memories are skewed by the presentation of new information

after an initial event. This effect has been tested in a multitude of ways, including testing words

lists, pictures, colors, and change blindness. Socially, the misinformation effect has been used to

explain the inaccuracies of eyewitness testimony. The current judicial system relies on the

fallible memory of people and has wrongfully imprisoned numerous innocents. The purpose of

this research is to show the misinformation effect is not a problem with memory storage and

retrieval, but rather a product of selective stimulus control.

ii

Copyright 2014

by

Kelly M. Tait

iii

ACKNOWLEDGEMENTS

First, I would like to acknowledge my advisor Dr. Jesus Rosales-Ruiz. Thank you for

putting up with me and my ideas throughout the years. I appreciate how you have stuck by me

and that I got to finish out my career at UNT researching with you. I would also like to thank Dr.

Jonathan Pinkston and Dr. Manish Vaidya for their insight and commentary throughout this

process. I would like to the BEHV 2300 teaching fellows and their classes for letting me come

and collect data. Lastly, I would like to thank my family for always encouraging me and having

faith in my abilities and knowledge.

iv

TABLE OF CONTENTS

Page ACKNOWLEDGEMENTS ........................................................................................................... iii LIST OF TABLES ...........................................................................................................................v LIST OF FIGURES ....................................................................................................................... vi INTRODUCTION ...........................................................................................................................1 METHODS ......................................................................................................................................7

Participants ...........................................................................................................................7

Setting ..................................................................................................................................7

Materials ..............................................................................................................................7

Data Collection ....................................................................................................................8

Procedures ............................................................................................................................8

Practice Set...........................................................................................................................8

Experiment Set .....................................................................................................................9

Design ................................................................................................................................10 RESULTS ......................................................................................................................................13 DISCUSSION ................................................................................................................................21 CONCLUSION ..............................................................................................................................26 APPENDIX A. INFORMED CONSENT FORM .........................................................................34 APPENDIX B. DATA COLLECTION FORM ............................................................................37 REFERENCES ..............................................................................................................................39

v

LIST OF TABLES

Page 1. Experimental Design for Groups 1 and 2 ..........................................................................11

2. Experimental Design for Groups 3 and 4 ..........................................................................12

3. Total Students Who Selected the Decoy ............................................................................19

4. Group 1’s Breakdown of Students Responding in Each Condition ...................................27




8. Overall Breakdown of Students Who Selected the Decoy in the Original + Decoy Discrimination Condition Across All Groups....................................................................30

vi

LIST OF FIGURES

Page

1. Group 1’s percentage of students’ responding per answer in each condition ....................13




5. Distribution of students who were misinformed in each set ..............................................20

6. Set 1. Face manipulation ....................................................................................................31

7. Set 2. Hair manipulation ....................................................................................................31

8. Practice set .........................................................................................................................32

9. Experimental Set 1 .............................................................................................................33

1

INTRODUCTION

The criminal justice systems uses a variety of means to arrest and convict suspected

criminals. Eye witness testimony, forensic evidence, and DNA testing are common methods that

provide evidence against perpetrators. While forensic evidence and DNA testing follow strict

scientific methodology to decipher evidence, eyewitness testimony relies on eyewitnesses’

memories of events (Grisham, 2009; Lindsay, 1990; Lindsay, Allen, Chan, & Dahl, 2004;

Loftus, 1986; Loftus, 1979a; Loftus, 1975; Szpitalak & Polczak, 2013; Wells & Loftus, 2003;

Wells & Olson, 2003). Social research has identified two variables that can be effect eyewitness

testimony. One is an estimator variable, which is defined as, “those that cannot be controlled by

the criminal justice system” (Grisham, 2009). This includes the location of the crime, visibility,

the distance between the witness and perpetrator, and the amount of stress the witness

experienced. The second type is a system variable and is defined as a variable that the “criminal

justice system can and should control” (Grisham, 2009). These include how evidence is gathered

and, of the interest to this study, how suspects are identified by eyewitnesses.

These two variables describe the circumstances surrounding the event and ways

eyewitnesses are asked to remember the event. A third variable investigators have taken an

interest in, is how memory recollection can be effected after the initial event. Cognitive theory

states that memories are retrievable or accessible but they can become skewed and distorted

when new information is presented. As a consequence of these distortions, the new memories

become more accessible than the originals. In the context of the criminal justice system, a

witness may at one point be able to accurately recall a suspect, but after time passes and they are

exposed to new information through media, conversations, interviews, etc., the original

memories can become altered (Loftus, 1974; Ross, Ceci, Dunning, & Toglia, 1994; Thomas,

2

Bulevich, & Chan, 2010). When eyewitnesses remember the new information instead of the

original, this change in memory has been labeled the misinformation effect.

Loftus (2005) defined the misinformation effect as “the change (usually for the worse) in

reporting that arises after receipt of misleading information” (p. 391). This effect occurs when

new information about the event is presented following the initial event. Under these

circumstances, eyewitnesses, when asked to recall the event, are more likely to recall the new

memories about the event instead of the original one. For example Loftus, Miller, and Burns

(1978) researched how exposure to misinformation between an initial event and subsequent

remembering of the event would evoke an incorrect memory. In this study, participants were

shown a series of slides depicting an auto-pedestrian accident. The critical slides in this series

were a picture of a red car at a T-intersection at a stop sign or at a yield sign. Half the

participants were shown the stop sign and the half shown the yield sign. After reviewing all 30

slides, participants were given a questionnaire about the pictures they had just viewed. Half the

subjects were given a test with the critical question reading, “Did another car pass the red Datsun

while it was stopped at the stop sign?” The other half had the same question only “stop sign” was

replaced with “yield sign.” Following the questionnaire participants were given a short, non-

related task. After the delay, participants were shown 15 pairs of slides from the series and asked

to select which of the two they had previously seen in the original slide show. Forty-one percent

of the participants who had read the new information about the yield sign, incorrectly identified

that slide with the yield sign instead of the stop sign. The new and incorrect question had

successfully implanted a new memory. So, when participants were asked to identify the correct

picture, the new, misinformed memory was recalled instead of the original one.

3

A second example of the misinformation effect was shown in a more recent, yet

unpublished, experiment completed by Cook, Kwak, Hoffman, and Loftus (2009). In this

experiment, the effect of misinformation on facial recognition were tested. Participants studied a

series of faces during a study phase. After the study phase, they were shown two faces and asked

to select which face matched the face from the study phase. One of the faces bore no

resemblance to the original, while the second picture, a decoy, was similar to the original in the

sense that one feature of the face was the same while another had been changed (e.g. the hair

remained the same, and the facial features were changed). Since one picture was definitely not a

match, the participants were more likely to choose the decoy picture as correct. In the last phase,

participants were shown the original face and the decoy and were asked to choose which picture

matched the original study face. There were participants who chose the decoy as a match to the

original, instead of the original itself. Like the Loftus et al. (1978) experiment, the researchers’

explanation was that selecting the decoy created a new memory and was more likely to be

recalled than the original.

The misinformation effect has also been researched in other ways:

An increase of alternate stimuli and allowing second guessing (Wright, Varley, & Belton,

1996)

Giving blatantly false information (Loftus, 1979; Loftus et al., 1978)

Change blindness which is defined as large changes occurring behind distracting stimuli,

and identifying perpetrators (Davies & Hine, 2007; Levin et al., 2002; Nelson et al.,

2001; Simons & Levin, 1998)

Increasing delays between study phases, misinformation, and testing phases (Loftus,

Miller, & Burns, 1978)

Identifying people in pictures and videos (Cook et al., 2009; Davies & Hines, 2007;

Halvorsen, 2012)

4

Variations of recognition including color, labels, names (Carmichael, Hogan, & Walter,

1932; Curran 2001; Daniel, 1972; Loftus, 1979; Loftus, 1977; Loftus, 1975; McCloskey

& Zaragoza, 1985; Wiese & Daum, 2006)

Word lists (Curran & Cleary, 2003; Deese, 1959a; Deese, 1959b; Roediger &

McDermott, 1995)

Yes/no, old/new discriminations, two stimuli forced choice (Allen, Wilding, & Rugg,

1998; Belli, 1989; Cook, et al., 2009; Curran, Schacter, Johnson, & Spinks, 2001;

Halvorsen, 2012; Rugg & Coles, 1995; Tversky & Tuchin, 1989; Weise & Daum, 2006)

All together the results of this research suggest that certain manipulations can change a

person’s memory. Memories can be supplemented, added to, altered, and transferred (Loftus,

1979; Loftus, 1975; Loftus et al., 1978).

The misinformation effect, however, does not happen to the majority of subjects in these

experiments. Consequently, the misinformation effect is explained as a product of individual

variables such as susceptibility and misattribution. A person can be susceptible to new memories

for a variety of reasons: age, working memory capacity, personality traits, socio-economic status,

race, and IQ (Ceci, Ross, & Toglia, 1987a). Misattribution occurs when a subject has incorrect

information but is unable to determine the differences between the original memory and who or

what implanted the new information. In addition, explanations of the misinformation effect

appeal to cloudy memories and weak traces of the original memory. A second explanation of the

misinformation effect has been described as a product of equivalence.

One assumption in all the research cited is that all the relevant information and features

acquired stimulus control during the initial exposure. However this is not always the case, so a

third explanation could be that participants may be under the control of fewer stimulus

dimensions. An example of selective attention in facial identification was shown in Duarte and

Baer (1997). They taught students to name certain hairstyles and then learn to name the same

hairstyles with an added face. The hair was established as the initial discriminative stimulus and

5

students either learned the whole face gradually or suddenly. Once the entire face was learned,

the hair was removed and students were asked to name just the face. On average, students could

only correctly name 5 out of 16 faces when the hair was missing. Further exposure to the faces

led to an increase in correct responding.

Similar to Duarte and Baer (1997), the results of Loftus et al. (1978) might be due to

selective attention. The assumption was that either the stop sign or the yield sign was part of the

stimulus control established during the initial presentation of the scene. However, this might not

be the case for some participants. If the signs were not part of the initial control then it is

reasonable to expect that for these participants their initial stimulus control could be expanded by

the new information. However, if the stop and yield signs were part of the initial control then the

participants would be less likely to be misinformed since the new information would conflict

with the old information.

Likewise, and more relevant to the present study, Cook et al. (2009) assumed that all the

relevant features of the face would acquire stimulus control. Again, it is possible that for some

participants not all the features of the face acquired stimulus control during the initial event. For

example if the initial control involved only the hairstyle, then by forcing them to choose the

same hairstyle with a modified face would not have an effect on them (e.g. Duarte and Baer,

1997). That is, no misinformation happened since the subjects were not basing their choice on

facial features. For subjects for whom the facial features were relevant, changing the response

would be less likely.

In regards to selective stimulus control and memories, Sidman (2008) asks:

The question to be asking is not where do memories go? but rather what were the original

conditions under which the behavior took place? What stimulus control topographies

were established? Then to the extent that those conditions, those topographies, can be

reinstated, people will remember things _ the behavior will come back. (p. 131)

6

The purpose of the current experiment was to show that the misinformation effect is a

product of the stimulus control acquired during the initial event rather than implanted memories.

The current study hypothesized that, if a particular feature of the picture acquired stimulus

control during the initial study phase that other features could be manipulated and the participant

would select other pictures that had the same feature. For example, if a participant selectively

attended to the hairstyle, then other features could be manipulated; the participant would

continue to select the picture that had the same hair as the original, regardless of the change in

face. Also, similar results would be found if a participant selectively attended to the face.

Manipulating the hairstyle would have no effect so long as the face remained the same as the

original. By including a neither condition, this experiment wanted to show that a forced choice

condition does not function as a misinformation manipulation but rather that it strengthens the

stimulus control that was acquired during the initial event. The current experiment does not

intend to describe ways to predict what a person will remember but give another more behavior

analytic explanation as to why people select or remember the way that they do.

7

METHODS

Participants

All participants in this study were students enrolled in an Introduction to Behavior

Analysis undergraduate course at the University of North Texas (UNT). There were four groups

of participants. Group 1 had 29 students and met during a fall semester Monday night class that

met for three hours. Group 2 had 21 students and met during a spring semester Monday night

class for three hours. Groups 3 and 4 met during a spring semester on Monday, Wednesday, and

Friday for one hour, in two separate classes and had 18 and 26 students respectively. These

students represent the diverse population of persons who may be asked to review a police line-

up, give eyewitness testimony, interact with police sketch artists, look over criminal

photographs, etc. Participants represent a wide ethnic diversity with students identifying

themselves as African-American, Asian, Caucasian-White, Hispanic, and Middle Eastern. Their

ages averaged 21.7 and ranged from 18 to 52. None of the participants informed the researchers

about any physical or intellectual limitations. Students were not compensated in anyway and

there were no foreseeable risks. The only benefits described to students was that of furthering

research and literature.

Setting

Each group’s experiment was conducted in a classroom at UNT. The room held

approximately 45 seats, long tables with retractable computers, a main computer, a projector, and

projection screen.

Materials

Materials included an informed consent form each participant signed before research

began (Appendix A). Students also received a data collection packet (Appendix B). Additional

8

materials included two slide shows: a practice set and an experiment set. The practice set differed

from the experimental set by using shapes instead of faces. The pictures used in the experiment

slide show were selected from Cook et al. (2009) and were divided into two sets. Set 1 contained

one original face (H1F1), one decoy (H1F2), and one arbitrary face; the decoy had the same

hairstyle as the original but the face was different (Figure 6). Set 2 also had one original face

(H1F1), one decoy (H2F1), and one arbitrary face; the decoy had the same face as the original

but the hairstyle was different (Figure 7). Both a digital and analog clock were used as timers.

Data Collection

Data collection was two sets with three questions each. The first question in each set only

allowed participants to mark A or B. The remaining 2 questions for each set allowed participants

to mark A, B, or Neither.

A hardcopy key was made by the researcher marking each correct response for each

question in each set. The results were then input into a spreadsheet that highlighted correct

answers and controlling stimuli based upon the participant’s responding. For example, if a

participant incorrectly selected the decoy whose hair was the same as the original, then

spreadsheet would input “hair” into the cell. All data sheets were compared to the key. If any

changes had been made on the data sheet, for example crossed out answers, the final response

was counted.

Procedures

Practice Set

The practice set used arbitrary shapes including a lightning bolt and a heart (See Figure

8). Choices made during the practice sets were not recorded. For the discrimination slides, there

was a visual prompt on the top of each slide stating, “Pick the picture you saw in the study phase.

9

Mark A or B, or Neither (when appropriate)”. The researcher also gave a vocal verbal prompt,

“Which picture matches the original? A, B, or Neither (when appropriate)?”

The study slide showed a single shape with a visual prompt, “Study Phase” on the top of

the slide. Participants were given 10s to study the shape; after time elapsed the slide went black.

The second slide was a forced choice slide with two simple shapes; participants choose the

matching shape by selecting A or B. After 10s, the slide went black. The neither slide showed the

same two shapes as the forced choice slide, but now had the options A, B, or Neither. After 10s

the slide went black. The fourth slide showed the original and a second shape. Participants could

respond A, B, or Neither.

Students in Group 1 were exposed to the practice set before the experiment set. Groups 2,

3, and 4 were only exposed to the experiment set.

Experiment Set

Groups 1 and 3 were shown Set 1 first and then Set 2. Groups 2 and 4 were shown Set 2

and then Set 1 (See Figure 9).

Study slide. Both sets showed the original picture, H1F1, in the middle of the slide with

the words “Study Phase” on the top of the slide. After 10s the slide went black.

Forced choice slide. For Set 1, the forced choice slide showed the arbitrary picture on the

left side and the decoy, H1F2, on the right side. The arbitrary picture had the label A under it

while the decoy was labeled B. The visual prompt, “Pick the picture you saw in the study phase.

Mark A or B on the top of the slide. Set 2’s forced choice was set up in the same way with the

arbitrary picture on the left and the decoy, H2F1, on the right. The arbitrary was labeled A and

the decoy, B. The same visual prompt was located on the top of the slide. The researcher also

10

gave a vocal verbal prompt, “Which picture matches the original? A or B?” The slide went black

after all students had answered or 30s elapsed, whichever came first.

Neither slide. The neither slides were identical to the forced choice slides for both sets,

but now included the word Neither between the two pictures. The visual prompt at the top of the

slide now stated, “Pick the picture you saw in the study phase. Mark A, B, or Neither.” The

researcher also gave a vocal verbal prompt, “Which picture matches the original? A or B?” The

slide went black after all students had answered or 30s elapsed, whichever came first.

Original + decoy discrimination slide. For Set 1, H1F2 was on the left side of the slide

and labeled A, and H1F1 was on the right side of the slide and labeled B. The word neither was

still placed between the two pictures and the same visual prompt from the neither slide was used.

In Set 2, H2F1 was located on the left side of the slide and labeled A, and H1F1 was on the right

side and labeled B. The word Neither was positioned between the two pictures. The same visual

prompt from the neither slide was used. The researcher also gave a vocal verbal prompt, “Which

picture matches the original? A or B?” The slide went black after all students had answered or

30s elapsed, whichever came first.

Design

This study used an A-B-B-A design including a counterbalance for the conditions. In the

A phase, the face was the variable that was changed. In the B phase, hair was the variable that

was changed. Groups 1 and 2 were exposed to the forced choice condition first, then the neither

condition and, lastly, the original + decoy discrimination condition (See Table 1). Groups 3 and 4

were shown the neither condition first, then the forced choice, then the original + decoy

discrimination (See Table 2).

11

Table 1

Experimental Design for Groups 1 and 2

Group 1

Set 1 Pictures Response Options Correct Response

Study Phase H1F1

Forced Choice Arbitrary 1, H1F2 A, B B

Neither Arbitrary 1, H1F2 A, B, Neither Neither

Original + Decoy

Discrimination H1F2, H1F1 A, B, Neither B


Study Phase H1F1



Original + Decoy


Group 2


Study Phase H1F1



Original + Decoy



Study Phase H1F1



Original + Decoy


12

Table 2

Experimental Design for Groups 3 and 4

Group 3


Study Phase H1F1



Original + Decoy



Study Phase H1F1



Original + Decoy


Group 4


Study Phase H1F1



Original + Decoy



Study Phase H1F1



Original + Decoy


13

RESULTS

The graphs for each group show the percentage of student responding across all three

conditions in each set. On the x-axis are the different responses students emitted in each

condition. On the y-axis is the percentage of students who emitted each response.

Group 1

Set 1 in the forced choice condition, 96.5% of the students selected the decoy and 93.1%

selected the decoy for Set 2. In the neither condition, 58.62% participants selected the decoy for

Set 1, and 31.03% selected the decoy for Set 2. In the original + decoy discrimination, 31.03%

selected the decoy for Set 1, and 3.45% selected Set 2’s decoy. (See Figure 1)

Figure 1. Group 1’s percentage of students’ responding per answer in each condition.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Decoy No

Response

Arbitrary Decoy Neither Decoy Original Neither

Per

centa

ge

of

Par

tici

pan

t R

esp

ond

ing

Responses

Group 1

Hair

Face

Forced Choice Original + Decoy

DiscriminationNeither

14

In Set 1, eight students selected the decoy in all three conditions in Set 1. Two students

selected the decoy in the first two conditions and then selected neither in the final discrimination

in Set 1. One student gave no response in the forced choice condition, selected neither in the

neither condition, and then chose the decoy in the last discrimination. In Set 2, one student

selected the decoy in all three conditions.

Overall, there were nine occurrences of students selecting the decoy in each condition for

Group 1, eight in Set 1 and one in Set 2. Of those who were misinformed in Set 1 when the

hairstyle stayed the same and the face changed, none selected the decoy in all three conditions in

Set 2. The same is true for the one individual who chose the decoy in all three conditions in Set

2; he/she did not select the decoys in Set 1. (See Table 4)

Group 2

For both Sets 2 and 1, 95.24% of participants selected the decoy in the forced choice

condition. In the neither condition, 33% selected the decoy for Set 2, and 43% selected the decoy

for Set 1. In the original + decoy discrimination condition, 10% of participants selected the

decoy for Set 2 and 14% selected the decoy in Set 1. (See Figure 2)

15


One student selected in the decoy in all three conditions for Set 2. Also, one student

selected the arbitrary picture in the forced choice condition, neither in the neither condition, and

then selected the decoy in the original + decoy discrimination condition in Set 2. Two students

selected the decoy in all three conditions in Set 1. One student chose the decoy in the forced

choice condition, then selected neither in both the neither condition and the original + decoy

discrimination condition. One student selected the decoy in the first condition, chose neither in

the neither condition and then chose the decoy in the final condition.

For Set 1, 20/21 participants chose the decoy in the forced choice condition. Nine out of

20 continued to select the decoy in the neither condition. In the final condition, two out of nine

selected the decoy.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Decoy Arbitrary Decoy Neither Decoy Original

Per

cen

tage

of

Par

tici

pan

t R

esp

on

din

g

Responses

Group 2

Face

Hair

Forced Choice Original + Decoy

DiscriminationNeither

16

Overall, three participants chose the decoy in all three conditions for both Sets 2 and 1.

The two students who selected the decoy in all three conditions in Set 1 did not select the decoy

in all three conditions in Set 2. The student who selected the decoy in all three conditions in Set 2

did not select the decoy in all three conditions in Set 1. (See Table 5)

Group 3

In the neither condition, 66.67% of participants selected the decoy for Set 1 and 27.78%

selected the decoy for Set 2. In the forced choice condition, 94.44% selected the decoy for Set 1

and 77.78% selected the decoy for Set 2. In the original + decoy discrimination condition,

22.22% selected the decoy in Set 1 and 11.11% selected the decoy in Set 2. (See Figure 3)


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Decoy Neither Decoy No

Response

Arbitrary Decoy Original Neither

Per

centa

ge

of

Par

tici

pan

t R

esp

ond

ing

Responses

Group 3

Hair

Face

Neither Original + Decoy

DiscriminationForced Choice

17

Three students selected the decoy in all three conditions in Set 1. One student chose the

decoy in the neither condition, the arbitrary in the forced choice condition, and then selected

neither in the original + decoy discrimination condition in Set 1. In Set 2, one students selected

the decoy in all three conditions. One student selected neither in the first condition and then

selected the decoy in the remaining two conditions.

In Group 3, there were four students who selected the decoy in every condition. Three

students selected all decoys in Set 1, and only one student in Set 2. None of the students who

selected the decoy in one set, selected the decoy in the other set. (See Table 6)

Group 4

In the neither condition 19.23% of participants selected the decoy for Set 2 and 34.62%

selected the decoy in Set 1. In the forced choice condition, 96.15% chose the decoy in Set 2 and

92.31% selected the decoy in Set 1. In the original + decoy discrimination condition, 3.85%

selected the decoy in Set 2 and 23.08% selected the decoy in Set 1. (See Figure 4)

18


In Set 2, one student selected the decoy in all three conditions. Two students chose

neither in the neither condition, the decoy in the forced choice condition, and neither in the

original + decoy discrimination condition. Four students selected the decoy in all three

conditions in Set 1. Two students selected neither in the neither condition, the decoy in the

forced choice condition, and the decoy in the last condition. Also in Set 1, two students selected

neither in the neither condition, the decoy in the forced choice condition, the neither in the last

discrimination condition. (See Table 7)

Overall Results

Table 3 shows how many students selected the decoy in each condition across all four

groups.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Decoy Neither Decoy Arbitrary Decoy Original Neither

Per

centa

ge

of

Par

tici

pan

t R

esp

ond

ing

Responses

Group 4

Face

Hair

Neither Original + Decoy

DiscriminationForced Choice

19

Table 3

Total Students Who Selected the Decoy

Set 1

Selected the Decoy in: Forced Choice Neither

Original + Decoy

Discrimination

Group 1 (n= 29) 28 17 9

Group 2 (n= 21) 20 9 3

Group 3 (n= 18) 17 12 4

Group 4 (n= 26) 24 9 6

Selected the decoy in all three conditions

Group 1 8

Group 2 2

Group 3 3

Group 4 4

Total: 17

Set 2

Selected the Decoy in: Neither

Forced

Choice

Original + Decoy

Discrimination

Group 1 9 27 1

Group 2 7 20 2

Group 3 5 14 2

Group 4 5 25 1

Selected the decoy in all three conditions

Group 1 1

Group 2 1

Group 3 1

Group 4 1

Total: 4

There were 21 occurrences of students selecting the decoy in all three conditions across a

total of 94 participants (See Table 8). Figure 5 shows that out of the 17 students who selected the

decoys in Set 1, 100% only chose the decoy in all three conditions in Set 1. The four remaining

20

students selected the decoy in all three conditions in Set 2 and of those four participants 100%

only selected the decoy in Set 2. None of these participants selected the decoy in all three

conditions in Set 1.

Figure 5. Distribution of students who were misinformed in each set.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Misinformed

when Face is

manipulated

Of those

misinformed in

Set 1, also

misinformed in

Set 2

Misinformed

when Hair is

manipulated

Of those

misinformed in

Set 2, also

misinformed in

Set 1

Per

centa

ge

of

Par

tici

pan

t R

esp

ond

ing

Responses

Distribution of Percentage of Participants Misinformed Across Sets

Misinformed when Face is manipulated

Misinformed when Hair is manipulated

Set 1Set 2

21

DISCUSSION

Overall, 94 % of the participants selected the decoy in Set 1 in the forced choice

condition, and 91% in Set 2. In the neither condition, 51% selected the decoy in Set 1, and 27%

in Set 2. In the original + decoy discrimination condition, 23% selected the decoy in Set 1 and

only 6% selected the decoy in Set 2. Out of the 94 participants across all four groups, 21 could

be considered misinformed, that is, they selected the decoy in all three conditions: seventeen

participants in Set 1 and four participants in Set 2. Of the 17 who selected the decoy in Set 1,

when the hair remained unchanged but the face was manipulated, none of those participants were

misinformed in Set 2, where the hair changed but the face remained the same. Similarly, of the

remaining four participants who were misinformed in Set 2, when the face remained the same

but the hair changed, none of them were misinformed in Set 1, when the hair remained the same

but the face changed. These results replicated Cook, et al. (2009), Halvorsen (2012) and Loftus

et al. (1978) in that a portion of the subjects were misinformed. The results also replicated

Halvorsen’s (2012) effects of finding an increasing in discriminating between the original and

the decoy due to the addition of the neither option. In addition, the results showed that the

misinformation effect could be due to selective responding rather than memory manipulations.

Similarly to Cook et al. (2009) and Loftus et al. (1978) the majority of the participants in

this study were sensitive to the misinformation manipulation that is, they selected the decoy in

the forced choice condition. However, even though the majority selected the decoy in this

condition, their memories were not altered, as evidenced in the final original + decoy

discrimination condition. Thus, the question remains: what was the controlling variable at the

moment of choice? One possibility, is that the response was under negative stimulus control

(Carrigan & Sidman, 1992), meaning that since the arbitrary picture did not share any of the

22

same features as the original and there was no option to reject, some participants were forced to

select the decoy by default. Another explanation could be that the common features between the

original and decoy controlled the selection. A third possibility is that for some participants they

were selectively responding to one feature, for example the hair or the face. The features that

changed during the misinformation condition were irrelevant to the control. For example, if the

hairstyle in Set 1 had acquired stimulus control, then, when the decoy had the same hairstyle,

participants would select the decoy based on just the hair since the feature manipulated during

the misinformation condition was irrelevant to their response. For the participants that selected

the arbitrary picture or did not make any selection, it might be that they could distinguish that the

decoy was different from the original and selecting the arbitrary choice was a way to handle the

forced choice (see Carrigan & Sidman, 1992).

The addition of a “neither” response helped clarify the results from the forced choice

condition. By selecting “neither” participants who were previously forced to choose the decoy,

now could show they could discriminate between the decoy and the original. Interestingly, the

effects of adding a “neither” option were similar regardless if the condition was presented after

the study phase or right after the forced choice condition. This suggests that the addition of the

neither condition did not have any instructional effect and just served to reveal existing stimulus

control. Since the stimulus control was reversible across the forced choice and neither

conditions, this also suggests that the forced choice did not act as a misinformation manipulation

for the majority of participants. Nonetheless, 22% of participants appeared to be misinformed by

the forced choice condition. However, the differential responding of those participants between

Set 1 and Set 2 show that they were selectively attending to restricted set of features and not due

to misinformation presented in the forced choice.

23

The experimental conditions were designed in such a way to see if the misinformation

effect could be explained by selective stimulus control instead of changes in initial memories. As

previously mentioned, accounts of the misinformation effect assume participants attended to all

relevant stimuli during the event to be remembered. However, it might be that for the

misinformed participants the stimulus control was restricted to certain features of the original

picture and the results depended on which features were controlling their choices and what

features were manipulated in the forced choice condition. To show this interaction, this

experiment manipulated different features in different sets so that, for example in Set 1, the hair

remained the same while facial features were changed. If subjects chose the decoy because they

were under the control of just the hair, then they would not be misinformed in Set 2 when the

hair was manipulated. Results from this experiment show that the persons who were

misinformed when the hair was not manipulated were not misinformed when the hair was

manipulated. Similarly participants who were under the control of just the face were not

misinformed in Set 1 when the facial features were manipulated but were misinformed in Set 2

when the face stayed the same. In other words, these results suggest that the misinformation

effect did not take place during the forced choice condition and appearance misinformation is an

artifact of selective attention (see Duarte & Baer, 1997; Reynolds & Limpo, 1969; Rosales-Ruiz,

Pasley, & Potucek, 1992; Sidman, 2008). This does not mean that one cannot expand on an

original memory, meaning that a new feature could be added if that feature did not have prior

stimulus control. For example, the memory of a yield a sign could be added if the stop sign did

not have original control (Loftus et al., 1978)

Previous explanations of the misinformation effect have included weak memory traces

and implanted memories. Equivalence has also been discussed as an explanation (Challies, Hunt,

24

Garry, & Harper, 2009). Their research into the misinformation effect showed participants could

be taught certain geometric patterns and then be misinformed during later equivalence trials. In

the first phase, participants were taught various patterns using simple, colored shapes. In the

second phase, misinformation, in the form of combining the original stimuli and novel stimuli,

was presented. In the final recognition test, 8% of participants who had been exposed to the

misinformation manipulations, incorrectly labeled the misinformation stimuli as patterns they

had learned in the initial learning phase. The researchers explained that incorrectly identifying

the new patterns was not a product of misinformation from second phase, but rather that stimulus

equivalence could account for the incorrect identification. Even though the authors designed

their experiment to decrease the possibility of select variables within the patterns acquiring

control instead of the whole pattern itself, they still acknowledge the possibility of selective

stimulus control occurring for two of the misinformation manipulations (e.g. common colors).

The current study agreed with this possibility of selective attention and the results showed that

what appeared as misinformation could actually be attributed to participants’ selectively

attending to only one feature of the picture instead of all the relevant variables.

There are a few limitations to this study. One limitation was the inability to prevent

participants from changing their answers after each slide. A verbal prompt to only mark one

choice was given before the beginning of the session but the researcher did not police the room

to enforce this request. Further research could incorporate the use of response cards that record

only the first response and would not allow students to change their answers. Analyzing the

initial response may increase likelihood for supposed misinformation since there would be no

opportunity to change answers.

25

Another limitation to this study was the number of sets and gender bias in the stimuli.

Future research should explore combining the hairstyles and faces from Sets 1 and 2. For

example the decoy would have the hairstyle from Set 1 but the face(s) from Set 2. Furthermore,

each set should include multiple decoys. Other research could also incorporate pictures of

females, since only male faces were used in the current experiment. To analyze the specific

variables in stimulus control, future research could test select stimulus removal and presentation,

similar to Duarte and Baer (1997).

In previous research, delaying the opportunity to recall, decreased correct identification

from 10% up to 50% (Loftus, 1979). Loftus et al. (1978), reported the strongest misinformation

effects occur after long delays. Future research should explore the accuracy of selecting the

original picture when there are delays between sets and conditions, from minutes to days (Loftus

et al., 1978; Loftus, 1979).

26

CONCLUSION

The misinformation effect has been an influential part in the study of eyewitness

identification (Bui, 2014; Grisham, 2009) and the malleability of human memory as evidenced

by Loftus and Hunter (1989):

We believe that the misinformation effect is sufficiently pervasive and eventually may be

so highly controlled that we are tempted to propose a Watsonian future for the

misinformation effect: Give us a dozen healthy memories, well formed, and our specified

world to handle them in. And we’ll guarantee to take any one at random and train it to

become any type of memory we might select- hammer, screwdriver, wrench, stop sign,

yield sign, Indian-chief- regardless of its origin or the brain that holds it. (p. 103) (Loftus

& Hunter, 1989)

In contrast, this study provides a framework to show that the possibility of creating a Watsonian

future may not be as simple as researchers claim. In this study, when participants who could be

described as misinformed, their behavior instead could be as a product of the functional control

of a particular stimulus. It is not that these particular students are more susceptible to

misinformation but rather that they were attending to only a particular feature and not all the

elements of the picture. The current study hypothesized that memories are not changed in the

ways that have been previously described. Instead, when participants make inaccurate selections,

the effect is due to individual differences in the acquisition of stimulus control during the initial

event. If more aspects of the original event to be remembered obtain stimulus control, then there

is the potential for better accuracy for recalling those details at a later time. Control by a select

stimulus may be mistaken for the misinformation effect since the participant chose the decoy

instead of the original picture.

Table 4

Group 1’s Breakdown of Students Responding in Each Condition

Set 1- HAIR number that selected decoy % selected decoy

Forced Choice 28/29 96.55%

Neither Choice 17/28 60.71%

Original + Decoy Discrimination 8/17 47.06%

Set 2- FACE




Total occurrences of the misinformation effect 9/29 31.03%

Participants misinformed in Set 1 8/29 27.59%

Participants misinformed in Set 2 1/29 3.45%

Participants misinformed in only Set 1 8/8 100.00%


27

Table 5


Set 2- FACE number that selected decoy

% selected

decoy




Set 1- HAIR





Participants misinformed in Set 2 (all three conditions) 1/21 4.76%




28

Table 6


Set 1- HAIR number that selected decoy % selected decoy




Set 2- FACE


Forced 14/18 77.78%







.

29

Table 7


Set 2- FACE number that selected decoy

% selected

decoy




Set 1- HAIR


Forced 24/26 92.31%







Table 8

Overall Breakdown of Students Who Selected the Decoy in the Original + Decoy Discrimination

Condition Across All Groups

Total Participants who

Selected the Decoy

Participants who

Selected the Decoy in

Set 1

Participants who

Selected the

Decoy in Set 2

Group

1 9 8 1

Group

2 3 2 1

Group

3 4 3 1

Group

4 5 4 1

Totals 21 17 4

30

H1F1 H1F2 Arbitrary

Figure 6. Set 1. Face manipulation.

H1F1 H2F1 Arbitrary

Figure 7. Set 2. Hair manipulation.

31

Figure 8. Practice set. From left to right: study slide, forced choice slide, neither slide, original +

decoy discrimination slide.

32

Figure 9. Experimental Set 1. From left to right: study slide, forced choice slide, neither slide,

original + decoy discrimination slide.

33

34

APPENDIX A

INFORMED CONSENT FORM

University of North Texas Institutional Review Board

Informed Consent Form

Before agreeing to participate in this research study, it is important that you read and understand

the following explanation of the purpose, benefits and risks of the study and how it will be

conducted.

Title of Study: “A Stimulus Control Analysis of the Misinformation Effect“

Principal Investigator: Jesús Rosales-Ruiz, University of North Texas (UNT) Department of

Applied Behavior Analysis.

Student Investigator: Kelly Tait, University of North Texas (UNT) Department of Applied

Behavior Analysis.

Purpose of the Study: You are being asked to participate in a research study which involves

how we as humans are able to retain information when presented with complex pictures. We

seek to learn if there are specific parts of pictures that humans are more likely to remember.

Study Procedures: You will be asked to study some pictures and later attempt to recall those

pictures when presented to you again, indicating on your paper which is the correct match. The

process will take about 5 to 10 minutes of your time.

Foreseeable Risks: There are no foreseeable risks involved in this study.

Benefits to the Subjects or Others: This study is not expected to directly benefit to you,

however it may yield important information about factors that influence your perception of

visual stimuli. This is especially important in legal and judicial systems, and pertinent academic

fields. This study may further our understanding of memory, perception, and recall of visual

stimuli. This study may benefit the field of behavior analysis, specifically pertaining to the

literature of stimulus control.

Procedures for Maintaining Confidentiality of Research Records: The confidentiality of your

individual information will be maintained in any publications or presentations regarding this

study. The information gathered through the questionnaires and informed consent will be kept

separate in a locked box. Only the Principal Investigator and Student Investigator will have

access to the data containing your personal information. All identifying markers will be removed

from publications so you cannot be identified. No personal information can be obtained unless

both consent form and questionnaire are acquired. Your identifying marker will be a number at

the top of the questionnaire form which correlates to the consent form. The data will be kept until

it is no longer valid or a minimum of 3 years. At that point the Principal Investigator and Student

Investigator may destroy the documents by shredding and/or burning.

Questions about the Study: If you have any questions about the study, you may contact Kelly

Tait at _______. Put “Misinformation Effect” in the subject.

Review for the Protection of Participants: This research study has been reviewed and

approved by the UNT Institutional Review Board (IRB). The UNT IRB can be contacted at (940)

565-3940 with any questions regarding the rights of research subjects.

Research Participants’ Rights: Your signature below indicates that you have read or have had read to you all of the above and

that you confirm all of the following:

35

• Kelly Tait has explained the study to you and answered all of your questions. You have been

told the possible benefits and the potential risks and/or discomforts of the study.

• You understand that you do not have to take part in this study, and your refusal to participate or

your decision to withdraw will involve no penalty or loss of rights or benefits. The study

personnel may choose to stop your participation at any time.

• Your decision whether to participate or to withdraw from the study will have no effect on your

grade or standing in this course.

• You understand why the study is being conducted and how it will be performed.

• You understand your rights as a research participant and you voluntarily consent to participate

in this study.

• You have been told you will receive a copy of this form.

________________________ ______________________________

Printed Name of Participant Signature of Participant

Date_________________

For the Principal Investigator or Designee: I certify that I have reviewed the contents of this form with the subject signing above. I have

explained the possible benefits and the potential risks and/or discomforts of the study. It is my

opinion that the participant understood the explanation.

______________________________________ __________

Signature of Principal Investigator or Designee Date

36

37

APPENDIX B

DATA COLLECTION FORM

A Stimulus Control Analysis of the Misinformation Effect Master thesis study by Kelly Tait. University of North Texas (UNT) student.

Department of Behavior analysis.

Personal information

Participant number: ____________

Age: ___ Ethnicity: ______________ Gender: _______________

Indicate answers with A, B, or Neither (when applicable).

A corresponds with the picture to the left and B corresponds to the picture on the right.

Set 1

1. A B

2. A B Neither

3. A B Neither

Set 2

1. A B

2. A B Neither

3. A B Neither

During the study, on what feature(s) were you basing your decisions?

38

REFERENCES

Belli, R. F. (1989). Influences of misleading postevent information: Misinformation interference

and acceptance. Journal of Experimental Psychology: General, 118(1), 72-85.

Bui, L. (2014). Prince George’s police to transform photo lineups. Retrieved from:

http://www.washingtonpost.com/local/crime/prince-georges-police-transform-photo-

lineups/2014/02/09/e1513fe4-8e8a-11e3-b227-12a45d109e03_story.html.

Carmichael, L., Hogan, H. P., Walter, A. A. (1932). An experimental study of the effect of

language on the reproduction of visually perceived form. Journal of Experimental

Psychology, 15(1), 73-86.

Carrigan, P. F., & Sidman, M. (1992). Conditional discrimination and equivalence relations: A

theoretical analysis of control by negative stimuli. Journal of the Experimental Analysis

of Behavior, 58(1), 183-204.

Ceci, S. J., Ross, D. F. & Toglia, M. P. (1987a). Age differences in suggestibility: Narrowing the

uncertainties. (In S. J. Ceci, M. P. Toglia, & D. F. Ross (Eds.), Children's eyewitness

testimony (pp. 79-91). New York: Springer.

Challies, D. M., Hunt, M., Garry, M., & Harper, D. N. (2011). Whatever gave you that idea?

False memories following equivalence training: A behavioral account of the

misinformation effect. Journal of Experimental Analysis of Behavior, 96, 343-362.

Cook, M., Kwak, Hoffman, D., & Loftus, E. F. (2009). Misinformation about faces seen. Paper

presented at Council of Science Editors Annual meeting. Pittsburgh, Pennsylvania.

Retrieved from:

http://www.councilscienceeditors.org/files/presentations/2009/Loftus.pdf.

39

http://www.washingtonpost.com/local/crime/prince-georges-police-transform-photo-lineups/2014/02/09/e1513fe4-8e8a-11e3-b227-12a45d109e03_story.html

http://www.washingtonpost.com/local/crime/prince-georges-police-transform-photo-lineups/2014/02/09/e1513fe4-8e8a-11e3-b227-12a45d109e03_story.html

http://www.councilscienceeditors.org/files/presentations/2009/Loftus.pdf

Curran, T. (2001). Implicit learning revealed by the method of opposition. Trends in Cognitive

Science, 5, 503-504.

Curran, T., & Cleary, A. M. (2003). Using ERPs to Dissociate Recollection from Familiarity in

Picture Recognition. Cognitive Brain Research, 15,191-205.

Curran, T., Schacter, D. L., Johnson, M. K., & Spinks, R. (2001). Brain potentials reflect

behavioral differences in true and false recognition. Journal of Cognitive Neuroscience,

13(2): 201-216.

Daniel, T. C. (1972). Nature of the effect of verbal labels on recognition memory for form.

Journal of Experimental Psychology, 96(1), 152-157.

Davies, G., & Hine, S. (2007). Change blindness and eyewitness testimony. Journal of

Psychology, 141(4), 423-434.

Deese, J. (1959a). On the prediction of occurrence of particular verbal intrusions in immediate

recall. Journal of Experimental Psychology, 58(1), 17-22.

Deese, J. (1959b) Influence of interitem associative strength upon immediate free recall.

Psychological Reports, 5, 305-312.

Duarte, A. M. M., & Baer, D. M. (1997). Overselectvity in the naming of suddenly and gradually

constructed faces. In D. M. Baer & M. Pinkston (Eds.), Environment and Behavior (pp.

210-218). Colorado: Westview Press.

Grisham, J. (2009). Innocence project. Retrieved from http://www.innocenceproject.org

Halvorsen, L. I. (2012). Misinformation about the misinformation effect (Master’s Thesis).

Retrieved from:

http://digital.library.unt.edu/ark:/67531/metadc149598/m2/1/high_res_d/thesis.pdf

40

http://psych.colorado.edu/~tcurran/papers.html#pdf_13




Levin, D.T., Simons, D.J., Angelone, B.L., & Chabris, C.F. (2002). Memory for centrally

attended changing objects in an incidental real-world change detection paradigm. British

Journal of Psychology, 93, 289-302.

Lindsay, D. S. (1990). Misleading suggestions can impair eyewitnesses’ ability to remember

event details. Journal of Experimental Psychology, 16(6), 1077-1083.

Lindsay, D. S., Allen, B. P., Chan, J. C. K., & Dahl, L. C. (2004). Eyewitness suggestibility and

source similarity: Intrusions of details from one event into memory reports of another

event. Journal of Memory and Language, 50(1), 96-111.

Loftus, E. G. (2005). Planting misinformation in the human mind: A 30-year investigation of the

malleability of memory. Learning and Memory, 12, 361-366.

Loftus, E. F. (1986). Ten years in the life of an expert witness. Law and Human Behavior, 10(3),

241-263.

Loftus, E. F. (1979). The malleability of human memory. American Scientist, 67, 312-320.

Loftus, E. F. (1977). Shifting human color memory. Memory & Cognition, 5, 696-699.

Loftus, E. F. (1975). Leading questions and the eyewitness report. Cognitive Psychology, 7, 560-

572.

Loftus, E. F., & Hunter, H. G. (1989). Misinformation and memory: The creation of new

memories. Journal of Experimental Psychology: General, 118, 100-104.

Loftus, E. F., Miller, D. G. & Burns, H. J. (1978). Semantic integration of verbal information

into a visual memory. Human Learning and Memory, 4(1), 19-31.

Loftus, E. F., & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of

the interaction between language and memory. Journal of Verbal Learning and Verbal

Behavior, 13(5), 585-589.

41

McCloskey, M., & Zaragoza, M. (1985a). Misleading postevent information and memory for

events: Arguments and evidence against memory impairment hypotheses. Journal of

Experimental Psychology: General, 114(1), 1-16.

Nelson, K. J., Laney, C., Fowler, N. B., Knowles, E. D., & Loftus, E. F. (2011). Change

blindness can cause mistaken eyewitness identification. Legal and Criminological

Psychology, 16(1), 62-74.

Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not

presented in lists. Journal of Experimental Psychology: Learning, Memory, and

Cognition, 21(4), 803-814.

Rosales-Ruiz, J., Pasley, J., & Potucek, J. (1992). Interference in an identity-sorting task.

Experimental Analysis of Human Behavior Bulletin, 1(1), 6-10.

Ross, D. R., Ceci, S. J., Dunning, D., & Toglia, M. P. (1994). Unconscious transference and

mistake identity: When a witness misidentifies a familiar with innocent person. Journal

of Applied Psychology, 79(6), 918-930.

Sidman, M. (2008). Reflections on stimulus control. The Behavior Analyst, 31(2), 127-135.

Simons, D. J., & Levin, D. T. (1998). Failure to detect changes to people during a real-world

interaction. Pyschonomic Bulletin & Review, 5(4), 644-649.

Szpitalak, M., & Polczyk, R. (2013). Promoting eyewitness testimony quality: Warning vs.

reinforced self-affirmation as methods of reduction of the misinformation effect. Polish

Psychological Bulletin, 44(1), 85-91.

Thomas, A. K., Bulevich, J. B., & Chan, J. C. K. (2010). Testing promotes eyewitness accuracy

with a warning: Implications for retrieval enhanced suggestibility. Journal of Memory

and Language, 63, 149-157.

42

Tversky, B., & Tuchin, M. (1989). A reconciliation of evidence on eyewitness testimony:

Comments on McCloskey and Zaragoza. Journal of Experimental Psychology: General,

118, 86-91.

Weise, H., & Daum, I. (2006). Frontal positivity discriminates true from false recognition. Brain

Research, 1075, 183-192.

Wells, G. L., & Loftus, E. F. Eyewitness memory for people and events. In Goldstein, Alan M.

(Ed), (2003). Handbook of psychology: Forensic psychology, 11 (149-160). Hoboken,

NJ, US: John Wiley & Sons Inc.

Wells, G. L., & Olson, E. (2003). Eyewitness testimony. Annual Review of Psychology, 54, 277-

295.

Wright, D. B., Varley, S., & Belton, A. (1996). Accurate second guesses in misinformation

studies. Applied Cognitive Psychology, 10(1), 13-21.

43

A Stimulus Control Analysis of the Misinformation Effect/67531/metadc... · misinformation effect...

Documents

Transcript of A Stimulus Control Analysis of the Misinformation Effect/67531/metadc... · misinformation effect...