Directing Attention by Goal Priming Impacts Accuracy of Witness Memory and Influences Misinformation Effects

SOUTHERN CROSS UNIVERSITY 

I have read and understand the Rules Relating to Awards (Rule 3 Section 18 – Academic Misconduct Including Plagiarism) as contained in the SCU Policy Library. I understand the penalties that apply for plagiarism and agree to be bound by these rules. The work I am submitting electronically is entirely my own work.

 Learning and Memory (Session 1, 2020

Abstract

Eye-witness memory is sensitive to bias, inattentional blindness, decay and misinformation effects. One attention-bias is the weapon focus effect, whereby attention and memory are increased for the central aspect and hence decreased for the peripheral aspects of an incident. This study aims to create a weapon focus effect depicting a close to real-life situation featuring a context narrative about an illegal luggage operation on CCTV footage. Attention is directed either towards a perpetrator or the luggage. 148 students and staff from a regional university participated in the study. The results concluded that narrative instructions attuned attention towards a chosen component and that memory for that aspect was better than for those peripheral aspects that have not received attention. Misinformation effects have been found for the components not in focus, especially for the luggage when attention was directed towards the person. However, a strong weapon focus effect was not provoked; the focus on the object differed only marginally from the focus on the perpetrator in that test condition. Further suggestions include a more persuasive narrative about the danger of the item to be able to capture attention and distort memory away from the perpetrator.

Directing Attention by Goal Priming Impacts Accuracy of Witness Memory and Influences Misinformation Effects

The quality of eye-witness descriptions depends on perceptions and the memory of the observer. Attention is the necessary 'spotlight' to assign mental resources to an event in order to encode an experience in WM. The salience of the incident, context, pre-existing schemas and memories formed after the event all influence the final witness memory. The whole progression from encoding to retrieval is a malleable process, and eye-witness memory is therefore susceptible to bias and misinformation. The fact that memories are malleable is especially important in legal contexts of testimony.

Creating realistic conditions to examine the weapon focus effect in a laboratory is challenging. Fawcett et al. (2013) have pointed out in their review that there are significant differences between lab experiments, staged experiments and real-life events presenting weapons, and that these are hard to compare.

Our study intends to fill this gap and therefore developed video footage and a narrative context to instil a real-world situation, featuring a closed-circuit television CCTV controller who must identify the aspects of the crime and memorise them for later questioning. In two experiments, a separate primer is delivered as a context narrative prior to the event, guiding attention towards either a perpetrator or a potentially menacing object. The present study aims to investigate the effect of attentional focus on memory in the context of a witness observing a crime on CCTV footage. Specifically, the study will examine memory accuracy and misinformation effects to determine the outcomes of goal priming and the role of directed attention.

Attention is sometimes allocated by an unexpected stimulus or an emotional state that increases saliency, like a weapon creating arousal (Easterbrook, 1959) or an unusual object capturing interest (Carlson et al., 2016). The weapon focus effect has been defined by Loftus et al. (1987) and is described as the attentional bias given to a salient object in a specific context. As a consequence of the focussed attention towards a weapon, the perpetrator has not received the focus needed to create a stable visual memory.

Attention can also be intentionally directed towards a specific element via an expectation or one's motivation. When instructed to prioritise an item before encoding, the items in focus were remembered better than items that received no focus of attention (Hu et al., 2016). Attentional cueing (Most et al., 2005) allocates mental resources towards an explicitly defined goal and increases attention and memory performance for that target (Légal et al., 2017). Attention is, to some degree, selective and can be purposely focussed towards a person or an object of choice and enhance encoding in memory.

Nevertheless, like memory (Miller, 1955), attention is also limited (Scholl, 2001). These limitations in attention and memory allow for inattentional blindness and susceptibility to misinformation effects. Divided as well as focussed attention can cause inattentional blindness, where information that has not received attentiveness is overlooked (Hyman, 2016). Crime blindness occurs when people's attention is divided or selectively allocated towards a task; crimes happen in full sight, yet they are either not noticed or weakly remembered afterwards (Hyman et al., 2018; Rivardo et al., 2011). When experiencing a crime where the focus had been pointed towards a salient or dangerous object, peripheral elements, as well as the perpetrator, may not resist leading questioning or other influences (Loftus, 2005). Wherever there is ambiguity, limited attention or gaps in memory, there is also a risk for misinformation to corrupt or mislead a memory (Lane, 2006; Loftus, 2005; Rivardo et al., 2011). Decay over time further weakens a memory (Revpoš & Baddeley, 2006) and the process of retroactive interference, whereas information after an event is re-associated, weaves misinformation into existing memories. Shaping memories can go as far as implanting childhood memories that have never occurred by telling untrue stories (Loftus & Pickrell, 1995), or using instructions and false photographs (Wade et al., 2002). Attentional cueing provides a powerful tool to capture attention and improve performance on visual search tasks and memory, however leaving blind spots where attention was turned away.

The present research predicts that a weapon focus effect can be primed by directing attention as instructions are given towards an attentional goal (H1). As a consequence of these instructions, attention will be attuned towards the cued aspect and memory will be increased for that central aspect of a scene (H2), and result in a weaker memory for all peripheral aspects (H3). It is further hypothesised that the elements that are not the central aspect of the focus are more prone to misinformation (H4).

Methods

Participants

Participants were recruited via email from the student and staff population at a regional University. In total, 148 people were included in the final analysis, 73 Male (mean age 37 sd=15), 75 Female (mean age 34 sd=12). An additional 32 people participated but were not included due to incomplete data.

Materials

The instructions, context information, event video and witness questions were presented via an online survey using Survey Monkey software. There were two versions of this survey - one for the Person Focus Condition and one for the Bag Focus Condition. The two minute video was the same in each condition and consisted of a young man checking in some baggage at an airport, walking to a waiting area carrying 2 bags, putting a bag down near a garbage can, leaving after a short while (without this bag) to go to the departure gate where it was discovered he was missing a bag, the bag being located and the man being escorted by security. The surveys and video were identical apart from a paragraph setting the context. This paragraph (see Appendix A) primed the participants to direct their attention to either the man (e.g. they were told they might need to identify him later on) or the bag (the bag could contain stolen goods and might need to be described). There was also a short distractor video of a person hitting golf balls. Examples of questions and information provided to the participants can be found in Appendix B.

Procedure

Potential participants received an email inviting them to volunteer for the study. If they were interested in further information and to participate, they clicked on a web link in the email. This took them one of the versions of the study (randomly selected) that was presented using Survey Monkey software. After some information and consent details were viewed, the study began with the context paragraph directing participants to attend to the person or the bag. The video was then displayed, followed by some questions that introduced misinformation, then a distractor video and associated questions. A series of questions related to the target video were presented, alternating between the types of questions. The open format questions were about details related to the person and immediate surrounds (e.g. "Did the suspect have facial hair and if so describe the type of facial hair?") and the bag and immediate surrounds (e.g. "What colour was the bag that was left in the waiting area?"). The study ended with a question regarding the quality of the technical aspects of the video. For each question type a percentage correct score was calculated.

Design

This study was a mixed design with Attention Focus (Person, Bag) as the between subjects factor and Question Type (Person and surrounds, Bag and surrounds) as the within subjects factor. The dependent variables are general accuracy and misinformation acceptance as percentages.

Results

The general accuracy for overall memory in the person-focus condition (M=77.5) was lower than in the bag-focus condition (M=80). However, as illustrated in Figure 1., the man in the person-focus condition received 23% more accurate memory recall compared to the luggage. This was further indicated by significantly greater overall memory accuracy for the person (M=83.25) and lower memory accuracy for the bag (M=74.25).

In the bag-focus condition, again see Figure 1., the memory accuracy between the luggage and the person only differed by 5% in favour of the bag. Hence, even when the participants' attention was primed towards the luggage, the person must still have been taken notice of significantly and was therefore remembered quite accurately.

Figure 1. General accuracy for the person-focus condition is showing a 23% higher accuracy for the person compared to the bag and a 5% higher accuracy for the bag compared to the person in the bag-focus condition. The overall means for both conditions illustrate that the person received 9% more accurate memory than the bag. The error bars represent the 95% confidence intervals.

Conversely, these results indicate that goal priming towards the central aspect did result in better memory recall for the element in focus and worse recall for the peripheral aspects of the scene, confirming our second and third hypothesis (H2 and H3).

There was no significant difference found in susceptibility to misinformation between the person-focus condition (M=30) and the bag-focus condition (M=29). However, as illustrated in Figure 2., the bag is 18% more susceptible to misinformation in the person-focus condition (M=31.75). In contrast, the person is only half that amount (9%) vulnerable to misinformation (M=27.25) when the focus was primed towards the luggage.

Susceptibility to Misinformation

45
40
35
30
25
20
15
10

5 0

Person-Focus

Bag-Focus Person Bag

Means Person VS Bag

Figure 2. Susceptibility to misinformation for the person-focus condition is showing a 19% higher vulnerability to misinformation for the bag compared to the person and only a 9% higher vulnerability to misinformation for the person compared to the bag in the bag-focus condition. The overall means for both conditions illustrate that the bag received 4.5% more misinformation than the person. The error bars represent the 95% confidence intervals.

These results suggest that misinformation was more prone for peripheral aspects, in line with our fourth hypothesis (H4).

Discussion

The present study aimed to explore if a weapon focus effect can be induced via goal priming. To answer this question it is essential to outline some limitations regarding

the concept. The weapon focus effect is primarily observed when a weapon is in clear vision and not obstructed or hidden. Carlson et al. (2017) found that a concealed weapon did not create a weapon focus effect. This finding extends on unusual objects as well. An unusual object must be unexpected and challenging an existing schema to capture and hold the attention; only then is the memory for that object more reliable (Lew & Howe, 2017; Pickel 1999).

The present experiment failed at creating a distinct weapon focus effect. Three weaknesses explain this finding: the bag itself was relatively conventional, and the narrative about the bag being a dangerous object or a weapon was not particularly strong as it merely featured illegal items, and further its contents were concealed. It was as such a hidden weapon in question. It was therefore not surprising to find only a 5% higher accuracy for the memory of the bag compared to the person in the bag-focus condition. Even so, the first hypothesis (H1) is confirmed; however, the validity to create a weapon focus effect seems weak.

Nevertheless, as the results show, there is a significant difference between the memory accuracy concerning the person and the luggage depending on the goal priming, which confirms H2 and H3. Priming the attention through a context narrative had created a focus of attention towards the central aspect of the scene and has had a positive effect on memory, confirming our second hypothesis (H2). Furthermore, as attention has been manipulated towards a central aspect, there was evidence of inattentional blindness towards the peripheral aspects of the scene, confirming our third hypothesis (H3); specifically, in the person-focus condition.

Misinformation was high for the elements in the peripheral aspects, confirming our fourth hypothesis (H4). Interestingly, the bag has received twice as much misinformation as a peripheral aspect than the person. Why was the person given such an advantage in memory accuracy as well as misinformation susceptibility? Biggs et al. (2013) found that when participants themselves were holding a gun, memory accuracy of facial features was improved. It could be argued that controlling the CCTV footage is similar to holding a weapon, an advantage of knowledge, and attention is therefore guided towards the perpetrator. Curby and Gauthier (2007) determined that in visual short-term memory, faces are remembered better than objects. This is especially true in prolonged exposure conditions, as was the case in our experiment. We owe that effect to our inherent human bias to process living elements (Calvillo & Jackson, 2013) and social aspects of our environment; furthermore, ethnically similar faces are more effortlessly recognised than culturally different faces (Hourihan et al., 2012).

As observers set their attention towards particular information to be received, other information becomes irrelevant and cannot be captured and encoded in visual memory (Kreitz et al., 2016; Most et al., 2005). Mere exposure to stimuli in a scene does not necessarily create a lasting memory, as experiments with coins we use every day demonstrated; memories about details on these coins were weak (Martin & Jones, 1995).

Similarly to the phonological loop, whereby rehearsal of words creates a more reliable memory and enables consolidation in long-term memory for recall (Revpoš & Baddeley, 2006), the visuospatial sketchpad relies on deeper levels of processing to maintain an accurate memory. These deeper levels of processing for our experiment were dependent on the attentional set (Most et al., 2005), triggered through priming, as this influenced the salience of the event. Schema congruent items (Alba & Hasher, 1983) that omit further levels of processing might have hindered capturing intense enough attention and therefore deeper encoding for accurate recall and susceptibility to include misinformation. This has been the case with the (not so unusual) bag and the concealed 'weapon', in combination with the airport scene; therefore, not challenging existing schemas or creating unexpectedness. However, prioritising an element from the scene did allocate attention towards that component, boosting encoding, and subsequently, memory has been improved (Atkinson et al., 2018).

Further amendments to our experiment in order to capture a more significant weapon focus effect could, therefore, include a more persuasive narrative about the content of the bag. To enhance a relevant and dangerous object and directing the attention away from the person in the bag-focus condition, the story could introduce an explosive device, similar to a terrorist attack. Alternatively, the bag could contain rare living species that potentially could escape and cause chaos in the airport. Overall, the idea with a security guard watching CCTV footage, thus turning the participant into a real-life witness is nevertheless, with a few adjustments regarding the research questions, a promising approach.

References

Alba, J. W., & Hasher, L. (1983). Is memory schematic? Psychological Bulletin, 93(2),

203–231. https://doi.org/10.1037/0033-2909.93.2.203
Atkinson, A. L., Berry, E. D., Waterman, A. H., Baddeley, A. D., Hitch, G. J., & Allen, R.

J. (2018). Are there multiple ways to direct attention in working memory? Annals of the New York Academy of Sciences, 1424(1), 115–126. https://doi.org/10.1111/nyas.13634

Biggs, A. T., Brockmole, J. R., & Witt, J. K. (2013). Armed and attentive: holding a weapon can bias attentional priorities in scene viewing. Atten Percept Psychophys, 75(8), 1715-1724. https://doi.org/10.3758/s13414-013-0538-6

Calvillo, D. P., & Jackson, R. E. (2014). Animacy, perceptual load, and inattentional blindness. Psychon Bull Rev, 21(3), 670-675. https://doi.org/10.3758/s13423-013- 0543-8

Carlson, C. A., Dias, J. L., Weatherford, D. R., & Carlson, M. A. (2017). An Investigation of the Weapon Focus Effect and the Confidence–Accuracy Relationship for Eyewitness Identification. Journal of Applied Research in Memory and Cognition, 6(1), 82-92. https://doi.org/10.1016/j.jarmac.2016.04.001

Carlson, C. A., Pleasant, W. E., Weatherford, D. R., Carlson, M. A., & Bednarz, J. E. (2016). The Weapon Focus Effect: Testing an Extension of the Unusualness Hypothesis. Psychology Faculty Publications, 8. https://digitalcommons.tamusa.edu/psyc_faculty/8

Curby, K. M., & Gauthier, I. (2007). A visual short-term memory advantage for faces. Psychonomic Bulletin & Review, 14(4), 620–628. https://doi.org/10.3758/bf03196811

Easterbrook, J. A. (1959). The effect of emotion on the utilisation and organisation of behavior. Psychological Review, 66, 183–201. http://dx.doi.org/10.1037/h0047707

Fawcett, J. M., Peace, K. A., & Greve, A. (2016). Looking Down the Barrel of a Gun: What Do We Know About the Weapon Focus Effect? Journal of Applied Research in Memory and Cognition, 5(3), 257-263. https://doi.org/10.1016/j.jarmac.2016.07.005

Fawcett, J. M., Russell, E. J., Peace, K. A., & Christie, J. (2013). Of guns and geese: a meta-analytic review of the 'weapon focus' literature. Psychology, Crime & Law, 19(1), 35-66. https://doi.org/10.1080/1068316x.2011.599325

Frenda, S. J., Nichols, R. M., & Loftus, E. F. (2011). Current Issues and Advances in Misinformation Research. Current Directions in Psychological Science, 20(1), 20-23. https://doi.org/10.1177/0963721410396620

Hourihan, K. L., Benjamin, A. S., & Liu, X. (2012). A cross-race effect in metamemory: Predictions of face recognition are more accurate for members of our own race. J Appl Res Mem Cogn, 1(3), 158-162. https://doi.org/10.1016/j.jarmac.2012.06.004

Hu, Y., Allen, R. J., Baddeley, A. D., & Hitch, G. J. (2016). Executive control of stimulus- driven and goal-directed attention in visual working memory. Atten Percept Psychophys, 78(7), 2164-2175. https://doi.org/10.3758/s13414-016-1106-7

Hyman, I. E. (2016). Unaware Observers: The Impact of Inattentional Blindness on Walkers, Drivers, and Eye-witnesses. Journal of Applied Research in Memory and Cognition, 5(3), 264-269. https://doi.org/10.1016/j.jarmac.2016.06.011

Hyman, I. E., Wulff, A. N., & Thomas, A. K. (2018). Crime Blindness: How Selective Attention and Inattentional Blindness Can Disrupt Eye-witness Awareness and Memory. Policy Insights from the Behavioral and Brain Sciences, 5(2), 202–208. https://doi.org/10.1177/2372732218786749

Kreitz, C., Furley, P., Memmert, D., & Simons, D. J. (2016). The Influence of Attention Set, Working Memory Capacity, and Expectations on Inattentional Blindness. Perception, 45(4), 386-399. https://doi.org/10.1177/0301006615614465

Lane, S. M. (2006). Dividing attention during a witnessed event increases eye-witness suggestibility. Applied Cognitive Psychology, 20(2), 199-212. https://doi.org/10.1002/acp.1177

Legal, J. B., Chekroun, P., Coiffard, V., & Gabarrot, F. (2017). Beware of the gorilla: Effect of goal priming on inattentional blindness. Conscious Cogn, 55, 165-171. https://doi.org/10.1016/j.concog.2017.08.004

Lew, A. R., & Howe, M. L. (2017). Out of place, out of mind: Schema-driven false memory effects for object-location bindings. J Exp Psychol Learn Mem Cogn, 43(3), 404-421. https://doi.org/10.1037/xlm0000317

Loftus, E. F. (2005). Planting misinformation in the human mind: a 30-year investigation of the malleability of memory. Learn Mem, 12(4), 361-366. https://doi.org/10.1101/lm.94705

Loftus, E. F., & Pickrell, J. E. (1995). The Formation of False Memories. Psychiatric Annals, 25(12), 720–725. https://doi.org/10.3928/0048-5713-19951201-07

Loftus, E. F., Loftus, G. R., & Messo, J. (1987). Some facts about "weapon focus." Law and Human Behavior, 11(1), 55–62. https://doi.org/10.1007/BF01044839

Martin, M., & Jones, G. V. (1995). Danegeld remembered: Taxing further the coin head illusion. Memory, 3(1), 97–104. https://doi.org/10.1080/09658219508251499

Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological review, 63(2), 81. https://doi.org/10.1037/0033-295x.101.2.343

Most, S. B., Scholl, B. J., Clifford, E. R., & Simons, D. J. (2005). What you see is what you set: sustained inattentional blindness and the capture of awareness. Psychol Rev, 112(1), 217-242. https://doi.org/10.1037/0033-295X.112.1.217

Pickel, K. L. (1999). The influence of context on the "weapon focus" effect. Law and Human Behavior, 23(3), 299–311. https://doi.org/10.1023/a:1022356431375

Repovš, G., & Baddeley, A. (2006). The multi-component model of working memory: Explorations in experimental cognitive psychology. Neuroscience, 139(1), 5-21. https://doi.org/10.1016/j.neuroscience.2005.12.061

Rivardo, M. G., Brown, K. A., Rodgers, A. D., Maurer, S. V., Camaione, T. C., Minjock, R. M., & Gowen, G. M. (2011). Integrating inattentional blindness and eye- witness memory. North American Journal of Psychology, 13(3). https://www.researchgate.net/profile/Mark_Rivardo/publication/235978418_Integ rating_Inattentional_Blindness_and_Eyewitness_Memory/links/546e28480cf2b5f c17605c82/Integrating-Inattentional-Blindness-and-Eyewitness-Memory.pdf

Scholl, B. J. (2001). Objects and attention: the state of the art. Cognition, 80(1-2), 1–46. https://doi.org/10.1016/s0010-0277(00)00152-9

Wade, K. A., Garry, M., Read, J. D., & Lindsay, D. S. (2002). A picture is worth a thousand lies: using false photographs to create false childhood memories. Psychon Bull Rev, 9(3), 597-603. https://doi.org/10.3758/bf03196318

PART 2

Remembering induces forgetting during consecutive eye-witness questioning tasks

Hypothesis

Witness collaboration after an event can improve memory for discussed items, however, later questioning of one set of characteristics yet not another can produce retrieval-induced forgetting for those elements that have gone unmentioned (MacLeod, 2002). We hypothesised that in the final questioning phase the characteristics a witness has been questioned about in a first interrogation round (practised items, PI) would be recalled better at the cost of items that have not been asked (non-practised items, NPI); consequently, these will be forgotten. Memory for items not discussed during a collaborative narrative between witnesses, and also not mentioned during the first interrogation phase (neglected items, NI) will be particularly weak.

Methods

Participants

144 psychology students from Southern Cross University completed the study. Each participant was proficient in English and had normal respectively corrected to normal vision.

Design

The study was a between-within-subjects design consisting of two trials with two different sets of questions (set A, set B). The first dependent variable was the difference between practised items (PI) and not practised items (NPI) in percentage; the second dependent variable was the percentage of neglected items (NI) at the final questioning task.  

Materials

We chose the airport security-breach video sequence used by Gosper (2020). The CCTV footage intently depicted a real-life situation of a suspicious luggage operation. The study consisted of an introduction narrative, the video sequence, a mock discussion between the participant and a colleague (witness collaboration), a first distraction sequence, a first interrogation task, a second distraction sequence and the final questioning task. The collaboration sequence contained all 20 items to be remembered (PI and NPI): eight characteristics per person (suspect and accomplice), and four regarding the luggage. The first interrogation task contained either set A or set B of ten PI-questions (set A; first ten of 20 items and set B; second half of 20 items), distributed between two separate groups of randomly assigned participants to control for any apparent context-effects. The final questioning task contained 30 questions: ten PI-questions, ten NPI-questions and ten NI-questions.

Procedure

Participants were invited via email; a link directed them to the Survey Monkey website, and they individually concluded a set A or set B version (randomly assigned) of the study on an electronic device. First, the role and task of the participant was described: the participant was told to be a security guard at work, who is watching life CCTV footage and who needs to pay close attention, if anything unusual should arrive he will have to raise the alarm. Consequently, the video was shown, suddenly ending: "The recording broke. You're discussing with your colleague about what just happened." A new screen showed a written mock-dialogue between the participant and a colleague where they discussed the footage (in ex.: "Did you see that guy, at check-in 6, sunglasses, tall, dark hair, beard, maybe in his thirties, blue jeans, he checked in two bags, but ...", "I noticed another man, nervous, skinny, ..." etc.). The collaboration was followed by the first distraction task for 1 minute in the form of an image of the interior of an elevator with soft music. The next screen displayed the first phase of interrogation: ten PI-questions were shown one at a time (in.ex.: "How old was the suspect?", "What type of shirt was the accomplice wearing?"), the participant had 10 seconds to enter each answer. This procedure was followed again by a second distraction task. Finally, the concluding questioning task appeared where the participant had to answer all 20 items (PI and NPI) plus ten questions about neglected items (NI), randomly assorted on a single page. The participant was then thanked and debriefed about the experiment. PI and NPI answers were totalled and compared, NI answers were counted, and percentages were calculated.

References

MacLeod, M. (2002). Retrieval-induced forgetting in eye-witness memory: forgetting as a

consequence of remembering. Applied Cognitive Psychology, 16(2), 135–149. https://doi.org/10.1002/acp.782