I have aploaded 6 articles and one instruction file All articles that needed to be used for reference please if you have any question you may ask thank you

Chatting in the Face of the Eyewitness: The Impact of Extraneous Cell-Phone Conversation on Memory for a Perpetrator John E. Marsh University of Central Lancashire and University of Gävle Krupali Patel University of Central Lancashire Katherine Labonté Université Laval Emma Threadgold University of Central Lancashire Faye C. Skelton Edinburgh Napier University Cristina Fodarella, Rachel Thorley, Kirsty L. Battersby, Charlie D. Frowd, and Linden J. Ball University of Central Lancashire François VachonUniversité Laval Cell-phone conversation is ubiquitous within public spaces. The current study investigates whether ignored cell-phone conversation impairs eyewitness memory for a perpetrator. Participants viewed a video of a staged crime in the presence of 1 side of a comprehensible cell-phone conversation (meaningful halfalogue), 2 sides of a comprehensible cell-phone conversation (meaningful dialogue), 1 side of an incomprehensible cell-phone conversation (meaningless halfalogue), or quiet. Between 24 and 28 hr later, participants freely described the perpetrator’s face, constructed a single composite image of the perpetrator from memory, and attempted to identify the perpetrator from a sequential lineup. Further, participants rated the likeness of the composites to the perpetrator. Face recall and lineup identification were impaired when participants witnessed the staged crime in the presence of a meaningful halfalogue compared to a meaningless halfalogue, meaningful dialogue, or quiet. Moreover, likeness ratings showed that the composites constructed after ignoring the meaningful halfalogue resembled the perpetrator less than did those constructed after experiencing quiet or ignoring a meaningless halfalogue or a meaningful dialogue. The unpredictability of the meaningful content of the halfalogue, rather than its acoustic unexpectedness, produces distraction. The results are novel in that they suggest that an everyday distraction, even when presented in a different modality to target information, can impair the long-term memory of an eyewitness.

Keywords:distraction, cell phones, eyewitness memory, dialogue, halfalogue Personal accounts and perceptions of how an event under in- vestigation unfolds is a vital element in police investigations.

Indeed, the apprehension ofcriminal suspects is often aided bydescriptions of crimes and their perpetrators (Cutler & Kovera, 2010).

Accounts provided from eyewitness memory offer valuable informa- tion that can contribute to the arrest and conviction of offenders (Samaha, 2005), especially in cases wherein the “hard evidence” needed for a conviction is lacking (Ainsworth, 2002). Eyewitness memory is therefore a domain in which accuracy is crucial, and given its importance, investigations of the various factors that may moderate eyewitness error are vital. The auditory environment is just one component of a myriad of complex facets of information that one may experience when witnessing an event such as a crime. Little is known, however, about the influence of the auditory scene on what is perceived or encoded from complex visual scenes that one would experience when witnessing a crime. In this study we investigate the potential impact of extraneous cell-phone con- versations—an omnipresent facet of the auditory environment in public areas— on the capability of an eyewitness (a) to recall detailed and accurate information about a perpetrator’s face, (b) to construct a composite accurate likeness of that face, and (c) to identify the perpetrator from a sequential lineup of visually similar identities.

This article was published Online First June 12, 2017.

John E. Marsh, School of Psychology, University of Central Lancashire, and Department of Building, Energy and Environmental Engineering, University of Gävle; Krupali Patel, School of Psychology, University of Central Lan- cashire; Katherine Labonté, School of Psychology, Université Laval; Emma Threadgold, School of Psychology, University of Central Lancashire; Faye C.

Skelton, School of Applied Sciences, Edinburgh Napier University; Cristina Fodarella, Rachel Thorley, Kirsty L. Battersby, Charlie D. Frowd, and Linden J. Ball, School of Psychology, University of Central Lancashire; François Vachon, School of Psychology, Université Laval.

The research reported in this article was financially supported by British Academy Grant SG122309, awarded to John E. Marsh, Faye C. Skelton, and Charlie D. Frowd.

Correspondence concerning this article should be addressed to John E.

Marsh, School of Psychology, University of Central Lancashire, Darwin Building, Preston, Lancashire, United Kingdom PR1 2HE. E-mail:

[email protected] Canadian Journal of Experimental Psychology / Revue canadienne de psychologie expérimentale © 2017 Canadian Psychological Association 2017, Vol. 71, No. 3, 183–1901196-1961/17/$12.00http://dx.doi.org/10.1037/cep0000101 183 Within modern society, engaging in cell-phone conversation is known to have adverse consequences on cognition, particularly in relation to driver accuracy (Strayer & Johnston, 2001) and pedes- trian behavior (Stavrinos, Byington, & Schwebel, 2011). For a passive bystander, others’ halfalogues (halves of conversations such as a cell-phone conversation whereby only one speaker can be heard) are rated as more noticeable and intrusive than are dialogues (i.e., complete conversations, as in when one can hear both sides of the conversation;Monk, Fellas, & Ley, 2004).

Moreover, cognitive performance can be differentially affected by halfalogues and dialogues. For example,Emberson, Lupyan, Goldstein, and Spivey (2010; see alsoGalvan, Vessal, & Golley, 2013) found that ignoring a halfalogue compared with a dialogue produced disruption to performance on a visual monitoring (track- ing) task and a choice reaction task. Although the existing evi- dence has suggested that overhearing half of a cell-phone conver- sation is enough to reduce performance on a concurrent, attentionally demanding task, there has been no attempt to inves- tigate the potential impact of ignoring cell-phone conversations on the recall of complex visual information in more applied tasks such as following the witnessing of a (staged) crime.

Typically, existing work on distraction via background sound has found impairment of short-term memory (STM) for sequences of visually presented items (e.g.,Hughes, Vachon, & Jones, 2005), but no study has shown impairment of long-term memory (LTM) when sound is presented during the encoding of visual material.

Certainly, from what is known about auditory distraction, it should be the case that background sounds that cause attention to be withdrawn from the prevailing task will impair encoding of visual events and therefore the later ability to recall those events from LTM.

One type of auditory distraction has been attributed to atten- tional diversion and occurs when the sound draws the attentional focus away from the prevailing mental activity (such as when an unexpected acoustic deviation is detected; e.g., the “m” in the irrelevant sequence “kkkkkkkmkk”;Hughes, Vachon, & Jones, 2007). Another type of auditory distraction is attributable to interference-by-process (Jones & Tremblay, 2000). Essentially, performance impairment ensues when there is a conflict between processes engaged to perform the focal task and processes applied involuntarily to the sound.

According to the attentional diversion standpoint, overhearing half of a conversation during study could impair encoding and therefore later recall from LTM at test because attention is directed involuntarily toward the sound due to a “need-to-listen.” This need-to-listen is driven by the tendency to predict the semantic content of the inaudible half of the conversation (Monk et al., 2004;Norman & Bennett, 2014). Attentional diversion can also occur due to rudimentary processing of the acoustic features of the ignored speech (Hughes et al., 2007): The unexpected onset and offset of the voice within one side of a phone conversation could produce a violation of the expectancy of auditory events within the sound stream, causing a disengagement of attention away from the focal task and impoverished recall of visual events. This “atten- tional capture” produced by the unpredictable onsets and offsets of a cell-phone conversation would be synonymous with the finding that unexpected changes in the pattern of auditory stimulation (e.g., the “m” in the irrelevant sequence “kkkkkkkmkk”) impairs STM for a sequence of visually presented items (e.g.,Hughes et al., 2005,2007;Vachon, Hughes, & Jones, 2012).

Therefore, both the need-to-listen and attentional capture accounts suggest that distraction is produced via attentional diversion.

According to the interference-by-process view, only tasks that require retention of serial order information should be vulnerable to distraction via changing-state sound (i.e., sound sequences that demonstrate abrupt changes in their acoustic properties; e.g., “c t g u”;Beaman & Jones, 1997). However, in contrast to the distrac- tion produced by interference-by-process, attentional diversion effects occur regardless of the task processes involved (Hughes et al., 2007;Vachon, Labonté, & Marsh, 2016). Therefore, if a half-conversation produces an attentional diversion effect, then disruption should manifest in complex cognitive tasks regardless of whether it involves serial STM. Witnessing and remembering an event is an example of such a task: Witnesses encode complex visual and/or auditory information that must be maintained so that it may later be recalled. Any distraction during the event may prevent eyewitnesses from encoding details that would later help to retrieve information from LTM, impacting negatively on their memory for event and person details. The Current Study The current study’s primary aim was to determine whether a to-be-ignored halfalogue negatively impacts on the LTM of an eyewitness to a staged crime. Attention was manipulated during the encoding of the crime event. Participants witnessed a video of a staged crime, prior to which they were told to ignore one of the following that occurred during the video: a full conversation (meaningful dialogue), a cell-phone conversation (meaningful hal- falogue) in a language they spoke, a spectrally rotated cell-phone conversation (incomprehensible to the participant and hence a “meaningless halfalogue”), or no sound (quiet). Between 24 and 28 hr later, the same participants described the perpetrator’s face from the staged-crime video in as much detail as possible and constructed a computer-generated likeness of the perpetrator (a composite). Finally, the participants were presented with a sequen- tial lineup (cf.Steblay, Dysart, Fulero, & Lindsay, 2001) of nine static facial photographs that included the perpetrator and eight distractor faces that were similar to that of the perpetrator in overall visual appearance. For each facial photograph, the partic- ipants were required to rate on a scale of 1–7 how certain they were that the identity depicted was the person they witnessed in the staged-crime video they viewed the previous day. These tasks were selected due to their ready use within police investigation (Frowd et al., 2013). Following this initial wave of experimentation, a set of independent judges rated the similarity of composites generated in each of the conditions (meaningful dialogue, meaningful halfa- logue, meaningless halfalogue, and quiet) to the perpetrator.

Given the demonstrable effect that unexpected auditory stimu- lation can have on simple attentional tasks (Emberson et al., 2010) regardless of the processes that underpin performance of the pri- mary task (Hughes et al., 2007), it was expected that ignoring a halfalogue would result in greater distraction than would ignoring a dialogue (and witnessing the staged crime in quiet; e.g., Emberson et al., 2010). Within this setting, distraction could manifest via recall of fewer correct facial details about the perpe- trator, impaired ability to identify the perpetrator from the sequen- tial lineup, and the production of composites that bear weak 184 MARSH ET AL. resemblance to the perpetrator. It is important to note that our inclusion of a meaningless halfalogue offered an opportunity to tease out whether any unique distraction produced by the halfa- logue could be attributable to a need-to-listen, whereby the seman- tic properties of the task-irrelevant speech draws attention from the primary task (Monk et al., 2004;Norman & Bennett, 2014), or to attentional capture, whereby an unexpected physical change in the auditory environment (such as the sudden onset of speech) is responsible for the withdrawal of attention from the focal task (e.g.,Hughes et al., 2005,2007). Method Participants Ninety-six students at the University of Central Lancashire (71 female) between the ages of 20 and 31 years (M 23.5,SD 3.21) took part in the main empirical study. Participants were recruited via opportunity sample. All participants spoke English as their first language and reported normal (or corrected-to-normal) vision and normal hearing. Twenty-four participants were allo- cated to each of the four sound conditions in the experiment. Nine participants did not return for the second part of the study and were replaced. A further 20 participants (14 female) between 21 and 37 years of age (M 25.9,SD 4.9) were recruited for the rating phase.

Apparatus and Materials Four versions of the same video of a staged crime that differed only with regard to the auditory background were used. The versions consisted of the following: quiet, a meaningful halfalogue (one side of a cell-phone conversation between two female speak- ers presented in the participants’ native language), a meaningless halfalogue (the sound presented for the meaningless halfalogue but spectrally rotated to render it incomprehensible), and a meaningful dialogue (two sides of the same cell-phone conversation presented as meaningful halfalogue). The same cell-phone conversation was therefore used for both the meaningful halfalogue and the mean- ingful dialogue conditions, with the former being created by de- leting one of the speaker’s voices. In the halfalogue version, there were nine pauses that ranged between 1.4 and 7.7 s (M 3.14, SD 2.08). The video and the cell-phone conversation lasted for 1 minute, and the onset of this conversation coincided with the onset of the video. The video depicted a man in his early 20s entering a corner shop and attempting to steal money from an unoccupied cash register—which could not be forced open— before making good his escape with several packets of cigarettes.

The topic of the phone conversation was based on a BBC news article about the nation’s favorite children’s book and was digitally recorded and sampled with a 16-bit resolution at a sampling rate of 44.1 kHz using a broadcast quality Dictaphone in an anechoic chamber. Halfalogues were created by silencing the voice of one of the speakers within the auditory file. The spectrally rotated halfa- logue was created by spectrally inverting the speech recording around 2 kHz (as inScott, Rosen, Beaman, Davis, & Wise, 2009).

Spectrally rotating speech involves transforming the high- frequency energy into low-frequency energy and vice versa. Spec- trally rotated speech is almost identical to normal speech (Scott etal., 2009). For example, variations in sound pressure level (SPL) across time and the duration of pauses between words and sen- tences are fairly equal. However, rotated speech is meaningless because it is incomprehensible.

The four versions of the same video (with different audio backgrounds) were created by embedding the audio onto the video using Windows Live Movie Maker (Microsoft Corporation, Red- mond, WA). Both normal speech and rotated speech were pre- sented over stereo headphones at approximately 69 dB (LAeq) as measured with an artificial ear.

The computer program PRO-fit (Version 3.5; ABM Limited, Nottingham, UK) was used to generate the facial composites.

PRO-fit is a feature-based system that involves presenting the witness with facial features (e.g., hair, eyes, nose, mouth) that match the face that the witness has previously described (for an overview, seeFrowd et al., 2014). This stage is described in more detail in the Procedure section. Procedure In the first session, participants viewed a staged-crime video in the context of one of the four sound conditions that they were randomly allocated to with equal sampling. They were seated at a distance of approximately 60 cm from the PC monitor in a testing cubicle and wore headphones. They were instructed to ignore any background sound, to know that they would not be asked anything about the sounds during the experiment, and to focus on studying the video. Participants were asked to return between 24 and 28 hr later, but the nature of the second visit was not revealed at this time.

In the second session, we revealed that a composite of the perpetrator witnessed in the staged-crime video would be required.

Participants were told that the goal of creating the composite was to produce an accurate portrayal of the perpetrator’s face so that another person could recognize the face as such. Participants were told that they would first describe the appearance of the face and then construct a composite of it. They were also told that there was no time limit to complete the face composite construction proce- dure (for existing articles explaining the detailed procedure for undertaking the face-recall interview and PRO-fit construction, see, e.g.,Frowd et al., 2013). In brief, participants were asked to think back to the time when the perpetrator had been seen, visu- alize the face, and then to try to recall as much detail about it as possible without guessing. The experimenter wrote down informa- tion that the participants recalled in relation to the face in this free-recall format. Participants were then informed that a compos- ite would be constructed of the face using PRO-fit. The experi- menter entered details from the face-recall phase into the descrip- tion details of PRO-fit. This generated the different features for the described face. If participants were not satisfied with a feature, then its size or location was adjusted or it was exchanged for another feature. Once participants reported that the best likeness had been achieved, the face was saved to a disk as the composite.

Following completion of the composite, participants undertook the sequential lineup task. They were given a sequential presenta- tion of facial photographs of nine identities that comprised the target (perpetrator), eight of which were foils that resembled the target to some extent in overall appearance. Using a 7-point Likert scale ranging from 1 (guess)to7(certain), participants were asked 185 IMPACT OF CELL-PHONE CONVERSATION ON MEMORY to indicate the certainty with which they considered that each facial photograph was the same identity as the person they wit- nessed in the staged-crime video they had viewed. The order in which the facial photographs were presented was pseudorandom:

Although the foils were presented in a random order for each participant, the target was presented in either Position 4 or Position 5 within the sequence. Participants were reminded that there was no time limit to complete the sequential lineup task. The time taken to complete the face composite construction and sequential lineup task varied between 25 and 45 min.

Once all of the composites had been constructed, other partici- pants were asked to rate the likeness of each of the composites compared to a frontal shot of the target (perpetrator) using a 7-point Likert scale ranging from 1 (very-poor likeness)to7(very-good likeness). Participants provided ratings for 96 composites (the 24 composites generated from within each sound condition). Composites were presented individually, each one next to the photograph of the target on a page in an A4 booklet. The presentation order of the composites was random for each participant.

Design The main empirical study (compared to the composite rating task) employed a between-subjects design whereby the indepen- dent variable was sound condition, with four levels: quiet, mean- ingless halfalogue, meaningful halfalogue, and meaningful dia- logue. For the face-recall part of the study (usually undertaken as part of a cognitive interview), the dependent variable was facial descriptor type, which had three levels: correct details, incorrect details, and subjective details; see further explanation later). For the sequential lineup component of the task, the independent variable was identity and had two levels: target (i.e., perpetrator) or foil, and the dependent variable was the confidence rating given to the target face and the mean rating given to the eight foils (col- lapsed). Finally, for the set of participants who independently rated the similarity of the composites to the target, the design was fully repeated measures, whereby the within-subject factor was sound condition (again quiet, meaningless halfalogue, meaningful halfa- logue, and meaningful dialogue) and the dependent variable was the similarity of each composite to the target rated on a scale of 1–7 (described earlier). Results Verbal Recall The quality of the face descriptions given by the participants within each sound condition was analyzed by two individuals.

Following the procedure used byMeissner, Brigham, and Kelley (2001), a correct description was generated by the two raters for the perpetrator’s face, and a decision was reached between the two raters as to which details would be classed as correct. Details in the descriptions were coded as correct, incorrect, or subjective. Sub- jective details were those that could not be verified directly (e.g., inferences about personality, or similarity to a well-known celeb- rity or family member). Interrater agreement was high, Cohen’s (72) .87,p .001 (Cohen, 1988). Contradictory scorings were resolved through discussion. The mean number of correct and incorrect features listed per condition can be seen inFigure 1. Themean number of correct descriptors provided was lower in the meaningful halfalogue condition compared to the meaningless halfalogue, meaningful dialogue, and quiet conditions. No differ- ence between means was apparent for incorrect descriptors. Only five details were classified as subjective descriptors across all four conditions, and because of this, subjective descriptors were ex- cluded in the further analysis.

A 4 (sound condition: meaningful dialogue vs. meaningful halfalogue vs. meaningless halfalogue vs. quiet) 2 (facial descriptor type: correct response vs. incorrect response) mixed factor analysis of variance (ANOVA) carried out on the mean number of face descriptors recalled revealed a main effect of facial descriptor type,F(1, 92) 47.70,MSE 6.61,p .001, with more correct than incorrect descriptors recalled ( p2 .34), but no main effect of sound condition,F(3, 92) 2.09,MSE 2.62,p .11, p2 .06. The Facial Descriptor Type Sound Condition interaction was significant,F(3, 92) 2.80,MSE 6.61,p .043, p2 .084. A simple-effects analysis (least significant difference [LSD]) revealed that correct facial de- scriptors were more frequent than incorrect facial descriptors for the quiet condition (p .001), meaningful dialogue condi- tion (p .001), and meaningless halfalogue condition (p .001) but not for the meaningful halfalogue condition (p .35).

Moreover, correct descriptors were less frequent in the mean- ingful halfalogue condition compared with the quiet condition (p .004), meaningful dialogue condition (p .012), and meaningless halfalogue condition (p .005). There was no difference between the means for the quiet and meaningless halfalogue conditions (p .95), quiet and meaningful dialogue conditions (p .70), and meaningless halfalogue and mean- ingful dialogue conditions (p .75). Moreover, there was no difference between conditions with respect to the frequency of incorrect information provided (p .1 for all comparisons).

Therefore, a to-be-ignored halfalogue, provided it is meaning- ful, presented during the witnessing of the staged-crime video diminished the quality of face description given the next day. Figure 1.Mean number of face descriptors recalled as a function of descriptor type and sound condition. Error bars represent the standard error of the mean. p .05. p .01. p .001. 186 MARSH ET AL. Sequential Lineup Task For the lineup task, the ratings reflecting the certainty that the identity was the same as the target in the video previously were addressed by comparing the mean rating given to the foil faces with the rating given to the target.Figure 2shows the mean certainty ratings for the foil identities (collapsed across identities) and the target for each of the four sound conditions. The confi- dence ratings were clearly greater for the target in the quiet, meaningful dialogue, and meaningless halfalogue conditions com- pared to the meaningful halfalogue condition. However, confi- dence ratings assigned to foil identities appears to differ little between conditions.

A 4 (sound condition) 2 (identity: target or foil) mixed- factorial ANOVA performed on mean confidence ratings revealed a main effect of identity, with higher confidence ratings for the target than for foils,F(1, 92) 250.12,MSE 1.91,p .001, p2 .73, but no main effect of sound condition,F(3, 92) 1.90, MSE 1.70,p .14, p2 .06. However, there was a significant Sound Condition Identity interaction,F(3, 92) 3.50,MSE 1.91,p .019, p2 .10. A simple-effects analysis (LSD) revealed that the mean confidence rating given to the target was lower in the meaningful halfalogue condition compared to the quiet condition (p .010), the meaningful dialogue condition (p .042), and the meaningfulness halfalogue condition (p .019). There was no significant difference between the quiet and meaningful dialogue conditions (p .58), quiet and meaningless halfalogue conditions (p .81), or the meaningful dialogue condition and the meaning- less halfalogue conditions (p .75). Therefore, a meaningful to-be-ignored halfalogue presented concurrently with the mock- crime video reduced the confidence with which the target was chosen from a lineup the next day. Composite Likeness Ratings Figure 3shows the means for the likeness scores given by the raters for the 24 composites within each of the four sound condi- tions. The mean values indicate that the raters considered that the composites generated in the quiet, meaningful dialogue condition, and meaningless halfalogue conditions looked more similar to the target face than did the composites generated in the meaningful halfalogue condition.

A one-way repeated-measures ANOVA demonstrated a signif- icant effect of sound condition on composite likeness,F(3, 57) 5.31,MSE .132,p .003, p2 .22. Planned repeated contrasts revealed that composites in the meaningful halfalogue condition bore less resemblance to the perpetrator than did those for the quiet condition (p .001), meaningless halfalogue condition (p .029), and meaningful dialogue condition (p .001). Additionally, those created in the meaningful dialogue condition were rated as better likenesses of the target face than were those created in the quiet condition (p .023; no other comparisons were significant).

Therefore, a meaningful to-be-ignored halfalogue presented con- currently with the mock-crime video resulted in facial composites that were rated poorer likenesses to the target.Figure 4show examples of the male target constructed in each of the sound conditions. Discussion To summarize, ignoring half of a cell-phone conversation, pro- viding it is meaningful, was shown to impair the long-term mem- ory (LTM) of the participant eyewitnesses. That the accuracy of eyewitness LTM—as measured through recall of facial descrip- tors, identification from a lineup, and composite accuracy—is susceptible to disruption via the presence of intermittent conver- sational background speech is important to acknowledge given the prominent role that eyewitnesses play in many criminal cases.

Composite images serve two purposes. On presentation within the media, they can generate leads from the general public to aid criminal investigations. They are also used as a reference from Figure 2.Mean confidence ratings as a function of sound condition in the context of the lineup task. These relate to whether the target or one of the foils was viewed earlier in the context of the mock-crime video. The mean ratings given to the eight foils are collapsed (1 guessand 7 certainthat the identity was seen earlier). Note therefore that a rating of 7 given to the target would essentially be a “hit,” whereas a rating of 1 given to the target would be a “miss.” Similarly, a rating of 1 given to a foil would be a “correct rejection,” whereas a rating of 7 to a foil would be a “false alarm.” Error bars represent the standard error of the mean. p .05. p .001. Figure 3.Mean likeness ratings awarded to the composites in the pres- ence of a photograph of the target as a function of sound condition (1 very poor likeness,7 very good likeness). Error bars represent the standard error of the mean. p .05. p .001. 187 IMPACT OF CELL-PHONE CONVERSATION ON MEMORY which criminal investigators can narrow likely suspects that may already be on file. Therefore, inaccuracies in eyewitnesses’ mem- ory—and subsequent composite quality— can potentially lead to false identifications (and arrests) and the pursuit of erroneous leads.

It is emerging that extraneous background speech can impair face memory in several ways. One way, for example, is through disruption of subvocal verbalization. It has become reasonably well accepted that spontaneous verbal codes are created for faces (Schooler, 2002). Indirect evidence that participants verbally re- hearse descriptions of faces within STM, and that such rehearsal ordinarily facilitates face recognition performance, comes from studies preventing subvocal verbalization by the use of articulatory suppression, a technique that requires participants to utter some repeated sounds (e.g., “ba ba ba ba”). Articulatory suppression impairs face recognition (Nakabayashi & Burton, 2008, Experi- ment 1;Nakabayashi, Burton, Brandimonte, & Lloyd-Jones, 2012; Wickham & Swift, 2006), whereas manual tapping—assumed to be as attentionally demanding as articulatory suppression without preventing verbalization— does not (e.g.,Nakabayashi & Burton, 2008, Experiment 3;Wickham & Swift, 2006). Whereas articula- tory suppression potentially eliminates the use of subvocal re- hearsal, extraneous changing-state speech (sound sequences that are acoustically changing [e.g., “c t g u”] compared to unchanging, steady-state speech [e.g., “c c c c”]) disrupts subvocal rehearsal due to processing conflict (seeJones, Madden, & Miles, 1992).

Consistent with the view that changing-state speech disrupts sub- vocal rehearsal and that subvocal rehearsal is used spontaneously to facilitate unfamiliar face learning,Marsh et al. (2017)have found that extraneous changing-state speech (randomly presented strings of letters), compared to steady-state speech (a string of the same letter repeated), presented during a 6-s exposure to a target face impairs recognition of that face from a lineup. However, that such interference is entirely independent of the semantic content of the speech suggests that the disruption is consistent with an interference-by-process view of distraction (Jones et al., 1992).

Here, the preattentive processing of the serial order of changes within sound interferes with the similar, deliberate process of subvocally rehearsing information derived from the visual modal- ity in serial order.

In the context of the current study, however, we favor an attentional diversion account (Hughes et al., 2007;Monk, Fellas, & Ley, 2004) over the disruption of subvocal rehearsal account for three reasons. First, participants did not know in advance that facerecall, composite construction, and lineup identification would be required subsequently. Therefore, the participants may not have rehearsed facial details explicitly. Second, perhaps counterintui- tively, the subvocal rehearsal process appears to utilize configural as opposed to featural information (Nakabayashi, Lloyd-Jones, Butcher, & Liu, 2012), which, according toSchooler (2002), involves information concerning the face’s global percept, includ- ing the spatial layout among its facial features. If disruption of subvocal rehearsal were the cause of face memory impairment, then it would appear quite counterintuitive that PRO-fit, a feature- based system (due to its requirement for recall of individual, isolated features and recognition of features in the context of the whole face) could capture the distraction effect. Third, since to- be-ignored meaningful dialogue speech—which presumably con- tains sufficient changing-state information to disrupt serial re- hearsal (Jones et al., 1992) and, in fact, more change than within halfalogues—failed to produce disruption, it is unlikely that the action of the meaningful to-be-ignored halfalogue speech is attrib- utable to the disruption of subvocal rehearsal.

Moreover, in the context of attentional diversion accounts (e.g., Hughes et al., 2007;Monk et al., 2004) the results of the experi- ment were unequivocal in providing support for the need-to-listen account of the halfalogue effect (Monk et al., 2004;Norman & Bennett, 2014) over an attentional capture account (cf.Hughes et al., 2005,2007). The halfalogue effect appeared only when the background speech material was meaningful. Because both the meaningful and meaningless (rotated) halfalogue speech were equated in terms of their acoustic complexity and temporal unpre- dictability, that only the meaningful halfalogue produces impair- ment refutes the idea that the halfalogue produces disruption due to the acoustic unexpectedness (and hence attentional capture) attrib- utable to the physical characteristics of sound (cf.Hughes et al., 2005). That the halfalogue effect is dependent upon the presence of semantic properties within the sound demonstrates that it is a form of distraction that differs from that attributable to acoustic unex- pectedness (Hughes et al., 2005,2007;Vachon, Hughes, & Jones, 2012). In the context of the current study, it appears that the meaningful halfalogue produces attentional diversion whereby the need-to-listen engendered by the tendency to want to predict or complete the missing part of the conversation causes an impover- ished encoding of details about the perpetrator, thereby impairing face recall and recognition. Although the task of face description, face construction, and target identification from a lineup are usu- ally carried out in this sequence in the real world, it is possible that Figure 4.Examples of the male target constructed in the four conditions of the experiment (displayed are those composites that have received the highest ratings for each sound condition). For copyright reasons, we are unable to reproduce the target photograph or stills from the video used in the experiment. 188 MARSH ET AL. these tasks may influence each other. For example, describing the target could have influenced the composite construction, and the composite construction may have influenced target identification in the lineup. Therefore, impoverished memory for the target produced by the meaningful halfalogue could have knock on effects at several loci within the procedures undertaken with the eyewitness.

Although it is perhaps intuitive that masking or otherwise in- terfering effects of additional environmental sounds such as voices may impede recognition and recall of a perpetrator’s voice (cf.

Stevenage et al., 2013), it is perhaps less intuitive that stimulation from a specific modality (auditory) should impair processing of information that is derived from another modality (visual). How- ever, the present findings unequivocally demonstrate that cell- phone conversation (meaningful halfalogue) breaks through selec- tive attention and impairs LTM even if participants know that the sounds contain no information that is relevant to the prevailing task (cf.Marsh et al., 2015) and therefore should be ignored.

To our knowledge the current results are novel in demonstrating that extraneous speech presented during encoding can produce adverse effects on LTM for complex visual information: the ap- pearance of a human face. Therefore, the findings illustrate the importance of considering the auditory environment when assess- ing the reliability of eyewitness memory. Moreover, these findings have implications far beyond the forensic context. Exposure to half of a conversation is a common occurrence that can impact nega- tively on one’s memory for complex visual information. Our results show that this irrelevant auditory information cannot sim- ply be ignored and as such has the potential to interfere with one’s processing of information in a wide range of daily activities.

Résumé Les conversations par téléphone cellulaire sont omniprésentes dans les espaces publics. La présente étude examine si le fait d’ignorer une conversation par téléphone cellulaire altère la mémoire d’un témoin oculaire par rapport au visage d’un malfaiteur. Les participants ont visionné une mise en scène de crime, sur vidéo, où l’on y présentait un côté d’une conversation par téléphone cellulaire compréhensible (milogue sensé), les deux côtés d’une conversation par téléphone cellulaire compréhensible (dialogue sensé), un côté d’une conversa- tion par téléphone cellulaire incompréhensible (milogue insensé), ou le silence. Entre 24 et 48 heures plus tard, les participants décrivaient librement le visage du malfaiteur, construisaient une image composite du malfaiteur a `partir de leur mémoire et tentaient d’identifier le malfaiteur dans une séance d’identification. Par la suite, les partici- pants devaient évaluer dans quelle mesure leur image composite ressemblait au malfaiteur. Le rappel du visage et l’identification lors de la séance d’identification étaient altérés chez les participants qui avaient été témoins de la scène de crime en présence d’un milogue sensé comparativement a `ceux qui étaient en présence d’un milogue insensé, d’un dialogue sensé ou du silence. De plus, les évaluations en termes de similarité ont montré que les composites construits après avoir ignoré les milogues sensés ressemblaient moins au malfaiteur que les composites construits par les participants en présence de silence, de milogue insensé ou de dialogue sensé. Le caractère im- prévisible du contenu sensé du milogue, plutôt que son imprévisibilité acoustique, engendre de la distraction. Les résultats sont innovateurs en ce sens qu’ils suggèrent qu’une distraction de tous les jours, mêmesi présentée sous une modalité différente pour cibler l’information, peut altérer la mémoire a `long terme d’un témoin oculaire.

Mots-clés: Distraction, téléphones cellulaires, mémoire d’un témoin oculaire, dialogue, milogue. References Ainsworth, P. B. (2002).Psychology and policing. Portland, OR: Willan.

Beaman, C. P., & Jones, D. M. (1997). The role of serial order in the irrelevant speech effect: Tests of the changing state hypothesis.Journal of Experimental Psychology: Learning, Memory, and Cognition, 23, 459 – 471.http://dx.doi.org/10.1037/0278-7393.23.2.459 Cohen, J. (1988).Statistical power analysis for the behavioral sciences.

Hillsdale, NJ: Erlbaum.

Cutler, B. L., & Kovera, M. B. (2010).Evaluating eyewitness identifica- tion. New York, NY: Oxford University Press.http://dx.doi.org/10 .1093/med:psych/9780195372687.001.0001 Emberson, L. L., Lupyan, G., Goldstein, M. H., & Spivey, M. J. (2010).

Overheard cell-phone conversations: When less speech is more distract- ing.Psychological Science, 21,1383–1388.http://dx.doi.org/10.1177/ 0956797610382126 Frowd, C. D., Jones, S., Fodarella, C., Skelton, F., Fields, S., Williams, A., . . . Hancock, P. J. B. (2014). Configural and featural information in facial-composite images.Science & Justice, 54,215–227.http://dx.doi .org/10.1016/j.scijus.2013.11.001 Frowd, C. D., Skelton, F., Hepton, G., Holden, L., Minahil, S., Pitchford, M.,...Hancock, P. J. B. (2013). Whole-face procedures for recovering facial images from memory.Science & Justice, 53,89 –97.http://dx.doi .org/10.1016/j.scijus.2012.12.004 Galván, V. V., Vessal, R. S., & Golley, M. T. (2013). The effects of cell phone conversations on the attention and memory of bystanders.PLoS One, 8,e58579.http://dx.doi.org/10.1371/journal.pone.0058579 Hughes, R. W., Vachon, F., & Jones, D. M. (2005). Auditory attentional capture during serial recall: Violations at encoding of an algorithm- based neural model?Journal of Experimental Psychology: Learning, Memory, and Cognition, 31,736 –749.http://dx.doi.org/10.1037/0278- 7393.31.4.736 Hughes, R. W., Vachon, F., & Jones, D. M. (2007). Disruption of short- term memory by changing and deviant sounds: Support for a duplex- mechanism account of auditory distraction.Journal of Experimental Psychology: Learning, Memory, and Cognition, 33,1050 –1061.http:// dx.doi.org/10.1037/0278-7393.33.6.1050 Jones, D., Madden, C., & Miles, C. (1992). Privileged access by irrelevant speech to short-term memory: The role of changing state.Quarterly Journal of Experimental Psychology A: Human Experimental Psychol- ogy, 44,645– 669.http://dx.doi.org/10.1080/14640749208401304 Jones, D. M., & Tremblay, S. (2000). Interference in memory by process or content? A reply to Neath (2000).Psychonomic Bulletin & Review, 7, 550 –558.http://dx.doi.org/10.3758/BF03214370 Marsh, J. E., Demaine, J., Bell, R., Skelton, F. C., Frowd, C. D., Röer, J. P., & Buchner, A. (2015). The impact of irrelevant auditory facial descrip- tions on memory for target faces: Implications for eyewitness memory.

Journal of Forensic Practice, 17,271–280.http://dx.doi.org/10.1108/ JFP-08-2014-0029 Marsh, J. E., Skelton, F. C., Vachon, F., Thorley, R., Frowd, C. D., & Ball, L. J. (2017). In the face of distraction:Irrelevant speech impairs person identification. Manuscript submitted for publication.

Meissner, C. A., Brigham, J. C., & Kelley, C. M. (2001). The influence of retrieval processes in verbal overshadowing.Memory & Cognition, 29, 176 –186.http://dx.doi.org/10.3758/BF03195751 Monk, A., Fellas, E., & Ley, E. (2004). Hearing only one side of normal and mobile phone conversations.Behaviour & Information Technology, 23,301–305.http://dx.doi.org/10.1080/01449290410001712744 189 IMPACT OF CELL-PHONE CONVERSATION ON MEMORY Nakabayashi, K., & Burton, A. M. (2008). The role of verbal processing at different stages of recognition memory for faces.European Journal of Cognitive Psychology, 20,478 – 496.

Nakabayashi, K., Burton, A. M., Brandimonte, M. A., & Lloyd-Jones, T. J.

(2012). Dissociating positive and negative influences of verbal process- ing on the recognition of pictures of faces and objects.Journal of Experimental Psychology: Learning, Memory, and Cognition, 38,376 – 390.http://dx.doi.org/10.1037/a0025782 Nakabayashi, K., Lloyd-Jones, T. J., Butcher, N., & Liu, C. H. (2012).

Independent influences of verbalization and race on the configural and featural processing of faces: A behavioral and eye movement study.

Journal of Experimental Psychology: Learning, Memory, and Cogni- tion, 38,61–77.http://dx.doi.org/10.1037/a0024853 Norman, B., & Bennett, D. (2014). Are mobile phone conversations always so annoying? The “need-to-listen” effect revisited.Behaviour & Infor- mation Technology, 33,1294 –1305.http://dx.doi.org/10.1080/ 0144929X.2013.876098 Samaha, J. (2005).Criminal justice. Belmont, CA: Wadsworth.

Schooler, J. W. (2002). Verbalization produces a transfer inappropriate processing shift.Applied Cognitive Psychology, 16,989 –997.http://dx .doi.org/10.1002/acp.930 Scott, S. K., Rosen, S., Beaman, C. P., Davis, J. P., & Wise, R. J. S. (2009).

The neural processing of masked speech: Evidence for different mech- anisms in the left and right temporal lobes.Journal of the Acoustical Society of America, 125,1737–1743.http://dx.doi.org/10.1121/1 .3050255 Stavrinos, D., Byington, K. W., & Schwebel, D. C. (2011). Distracted walking: Cell phones increase injury risk for college pedestrians.Jour-nal of Safety Research, 42,101–107.http://dx.doi.org/10.1016/j.jsr.2011 .01.004 Steblay, N., Dysart, J., Fulero, S., & Lindsay, R. C. L. (2001). Eyewitness accuracy rates in sequential and simultaneous lineup presentations: A meta-analytic comparison.Law and Human Behavior, 25,459 – 473.

http://dx.doi.org/10.1023/A:1012888715007 Stevenage, S. V., Neil, G. J., Barlow, J., Dyson, A., Eaton-Brown, C., & Parsons, B. (2013). The effect of distraction on face and voice recogni- tion.Psychological Research, 77,167–175.http://dx.doi.org/10.1007/ s00426-012-0450-z Strayer, D. L., & Johnston, W. A. (2001). Driven to distraction: Dual-task studies of simulated driving and conversing on a cellular telephone.

Psychological Science, 12,462– 466.http://dx.doi.org/10.1111/1467- 9280.00386 Vachon, F., Hughes, R. W., & Jones, D. M. (2012). Broken expectations:

Violation of expectancies, not novelty, captures auditory attention.Jour- nal of Experimental Psychology: Learning, Memory, and Cognition, 38, 164 –177.http://dx.doi.org/10.1037/a0025054 Vachon, F., Labonté, K., & Marsh, J. E. (2016). Attentional capture by deviant sounds: A noncontingent form of auditory distraction?Journal of Experimental Psychology: Learning, Memory, and Cognition. Ad- vance online publication.http://dx.doi.org/10.1037/xlm0000330 Wickham, L. H. V., & Swift, H. (2006). Articulatory suppression attenu- ates the verbal overshadowing effect: A role for verbal encoding in face identification.Applied Cognitive Psychology, 20,157–169.http://dx.doi .org/10.1002/acp.1176 Received May 26, 2016 Accepted August 24, 2016 E-Mail Notification of Your Latest CPA Issue Online! Would you like to know when the next issue of your favorite APA journal will be available online? This service is now available to you. Sign up at https://my.apa.org/portal/alerts/ and you will be notified by e-mail when issues of interest to you become available! Avis par courriel de la disponibilité des revues de la SCP en ligne! Vous voulez savoir quand sera accessible en ligne le prochain numéro de votre revue de la Société canadienne de psychologie préférée? Il est désormais possible de le faire. Inscrivez-vous a `https://my.apa.org/portal/alerts/ et vous serez avisé par courriel de la date de parution en ligne des numéros qui vous intéressent! 190 MARSH ET AL.