Visual search for changes in scenes creates long-term,incidental memory traces

Humans are very good at remembering large numbers of scenes over substantial periods of time. But how good are they at remembering changes to scenes? In this study, we tested scene memory and change detection two weeks after initial scene learning. In Experiments 1–3, scenes were learned incidentally during visual search for change. In Experiment 4, observers explicitly memorized scenes. At test, after two weeks observers were asked to discriminate old from new scenes, to recall a change that they had detected in the study phase, or to detect a newly introduced change in the memorization experiment. Next, they performed a change detection task, usually looking for the same change as in the study period. Scene recognition memory was found to be similar in all experiments, regardless of the study task. In Experiment 1, more difficult change detection produced better scene memory. Experiments 2 and 3 supported a “depth-of-processing” account for the effects of initial search and change detection on incidental memory for scenes. Of most interest, change detection was faster during the test phase than during the study phase, even when the observer had no explicit memory of having found that change previously. This result was replicated in two of our three change detection experiments. We conclude that scenes can be encoded incidentally as well as explicitly and that changes in those scenes can leave measurable traces even if they are not explicitly recalled.     

T1 difficulty affects the AB: manipulating T1 word frequency and T1 orthographic neighbor frequency

Colored target words were presented with distractor nonwords in a rapid serial visual presentation (RSVP) task. In Experiment 1, the attentional blink (AB) effect on T2 accuracy was larger when T1 was a difficult (low-frequency) word than when it was a high-frequency word. In Experiment 2 the effect of T1 frequency on the AB was replicated in a between-participants design, and the frequency of T1’s one-letter different neighbors (e.g., case, bare, for care) interacted with T1 frequency in its effects on T2 accuracy. Experiment 3 confirmed the effect of T1 frequency over 6 T1-T2 lags. The effects of T1 characteristics were sensitively assessed in the AB and were more consistent with resource depletion theories than control-process accounts.     

Dissecting stimulus–response binding effects: Grouping by color separately impacts integration and retrieval processes

In selection tasks, target and distractor features can be encoded together with the response into the same short-lived memory trace, or event file (see Hommel, 2004), leading to bindings between stimulus and response features. The repetition of a stored target or distractor feature can lead to the retrieval of the entire episode, including the response—so-called “binding effects.” Binding effects due to distractor repetition are stronger for grouped than for nongrouped target and distractor stimulus configurations. Modulation of either of two mechanisms that lead to the observed binding effects might be responsible here: Grouping may influence either stimulus–response integration or stimulus–response retrieval. In the present study we investigated the influences of grouping on both mechanisms independently. In two experiments, target and distractor letters were grouped (or nongrouped) via color (dis)similarity separately during integration and retrieval. Grouping by color similarity affected integration and retrieval mechanisms independently and in different ways. Color dissimilarity enhanced distractor-based retrieval, whereas color similarity enhanced distractor integration. We concluded that stimulus grouping is relevant for binding effects, but that the mechanisms that contribute to binding effects should be carefully separated.     

Remember to blink: Reduced attentional blink following instructions to forget

This study used rapid serial visual presentation (RSVP) to determine whether, in an item-method directed forgetting task, study word processing ends earlier for forget words than for remember words. The critical manipulation required participants to monitor an RSVP stream of black nonsense strings in which a single blue word was embedded. The next item to follow the word was a string of red fs that instructed the participant to forget the word or green rs that instructed the participant to remember the word. After the memory instruction, a probe string of black xs or os appeared at postinstruction positions 1–8. Accuracy in reporting the identity of the probe string revealed an attenuated attentional blink following instructions to forget. A yes–no recognition task that followed the study trials confirmed a directed forgetting effect, with better recognition of remember words than forget words. Considered in the context of control conditions that required participants to commit either all or none of the study words to memory, the pattern of probe identification accuracy following the directed forgetting task argues that an intention to forget releases limited-capacity attentional resources sooner than an instruction to remember—despite participants needing to maintain an ongoing rehearsal set in both cases.     

Identity and semantic negative priming in rapid serial visual presentation streams

In selective attention tasks, the efficiency of processing concurrently presented target and distractor stimuli in a given display is often influenced by the relationship these stimuli have with those in the previous display. When a to-be-attended target on a current trial (the probe trial) matches the ignored, non-target distractor on a previous trial (the prime trial), a response to the target is typically delayed compared with when the two stimuli are not associated. This negative priming (NP) phenomenon has been observed in numerous studies with traditional NP tasks presenting the target and distractor simultaneously in both the prime and probe trial couplets. Here, however, in four experiments using a mixture of stimulus types (letters, digits, English number words, and logographic Chinese number words), target and distractor stimuli were temporally separated in two rapid serial visual presentation (RSVP) streams instead of concurrently presented. The findings provide a conceptual replication and substantial extension of a recent study by Wong (Plos One, 7, e37023, 2012), and suggest that active suppression of irrelevant distracting information is a more ubiquitous form of cognitive control than previously thought.     

Effects of varying presentation time on long-term recognition memory for scenes: Verbatim and gist representations

Konkle, Brady, Alvarez and Oliva (Psychological Science, 21, 1551–1556, 2010) showed that participants have an exceptional long-term memory (LTM) for photographs of scenes. We examined to what extent participants’ exceptional LTM for scenes is determined by presentation time during encoding. In addition, at retrieval, we varied the nature of the lures in a forced-choice recognition task so that they resembled the target in gist (i.e., global or categorical) information, but were distinct in verbatim information (e.g., an “old” beach scene and a similar “new” beach scene; exemplar condition) or vice versa (e.g., a beach scene and a new scene from a novel category; novel condition). In Experiment 1, half of the list of scenes was presented for 1 s, whereas the other half was presented for 4 s. We found lower performance for shorter study presentation time in the exemplar test condition and similar performance for both study presentation times in the novel test condition. In Experiment 2, participants showed similar performance in an exemplar test for which the lure was of a different category but a category that was used at study. In Experiment 3, when presentation time was lowered to 500 ms, recognition accuracy was reduced in both novel and exemplar test conditions. A less detailed memorial representation of the studied scene containing more gist (i.e., meaning) than verbatim (i.e., surface or perceptual details) information is retrieved from LTM after a short compared to a long study presentation time. We conclude that our findings support fuzzy-trace theory.     

Normative accuracy and response time data for the computerized Benton Facial Recognition Test (BFRT-<Emphasis Type="Italic">c</Emphasis>)

We report normative data from a large (N = 307) sample of young adult participants tested with a computerized version of the long form of the classical Benton Facial Recognition Test (BFRT; Benton & Van Allen, 1968). The BFRT-c requires participants to match a target face photograph to either one or three of six face photographs presented simultaneously. We found that the percent accuracy on the BFRT-c (81%–83%) was below ceiling yet well above chance level, with little interindividual variance in this typical population sample, two important aspects of a sensitive clinical test. Although the split-half reliability on response accuracy was relatively low, due to the large variability in difficulty across items, the correct response times measured in this version—completed in 3 min, on average—provide a reliable and critical complementary measure of performance at individual unfamiliar-face matching. In line with previous observations from other measures, females outperformed male participants at the BFRT-c, especially for female faces. In general, performance was also lower following lighting changes than following head rotations, in line with previous studies that have emphasized participants’ limited ability to match pictures of unfamiliar faces with important variations in illumination. Overall, this normative data set supports the validity of the BFRT-c as a key component of a battery of tests to identify clinical impairments in individual face recognition, such as observed in acquired prosopagnosia. However, this analysis strongly recommends that researchers consider the full test results: Beyond global indexes of performance based on accuracy rates only, they should consider the time taken to match individual faces as well as the variability in performance across items.     

Postencoding cognitive processes in the cross-race effect: Categorization and individuation during face recognition

The cross-race effect (CRE) describes the finding that same-race faces are recognized more accurately than cross-race faces. According to social–cognitive theories of the CRE, processes of categorization and individuation at encoding account for differential recognition of same- and cross-race faces. Recent face memory research has suggested that similar but distinct categorization and individuation processes also occur postencoding, at recognition. Using a divided-attention paradigm, in Experiments 1A and 1B we tested and confirmed the hypothesis that distinct postencoding categorization and individuation processes occur during the recognition of same- and cross-race faces. Specifically, postencoding configural divided-attention tasks impaired recognition accuracy more for same-race than for cross-race faces; on the other hand, for White (but not Black) participants, postencoding featural divided-attention tasks impaired recognition accuracy more for cross-race than for same-race faces. A social categorization paradigm used in Experiments 2A and 2B tested the hypothesis that the postencoding in-group or out-group social orientation to faces affects categorization and individuation processes during the recognition of same-race and cross-race faces. Postencoding out-group orientation to faces resulted in categorization for White but not for Black participants. This was evidenced by White participants’ impaired recognition accuracy for same-race but not for cross-race out-group faces. Postencoding in-group orientation to faces had no effect on recognition accuracy for either same-race or cross-race faces. The results of Experiments 2A and 2B suggest that this social orientation facilitates White but not Black participants’ individuation and categorization processes at recognition. Models of recognition memory for same-race and cross-race faces need to account for processing differences that occur at both encoding and recognition.     

Are event-related potentials to dynamic facial expressions of emotion related to individual differences in the accuracy of processing facial expressions and identity?

Despite a wealth of knowledge about the neural mechanisms behind emotional facial expression processing, little is known about how they relate to individual differences in social cognition abilities. We studied individual differences in the event-related potentials (ERPs) elicited by dynamic facial expressions. First, we assessed the latent structure of the ERPs, reflecting structural face processing in the N170, and the allocation of processing resources and reflexive attention to emotionally salient stimuli, in the early posterior negativity (EPN) and the late positive complex (LPC). Then we estimated brain–behavior relationships between the ERP factors and behavioral indicators of facial identity and emotion-processing abilities. Structural models revealed that the participants who formed faster structural representations of neutral faces (i.e., shorter N170 latencies) performed better at face perception (r = –.51) and memory (r = –.42). The N170 amplitude was not related to individual differences in face cognition or emotion processing. The latent EPN factor correlated with emotion perception (r = .47) and memory (r = .32), and also with face perception abilities (r = .41). Interestingly, the latent factor representing the difference in EPN amplitudes between the two neutral control conditions (chewing and blinking movements) also correlated with emotion perception (r = .51), highlighting the importance of tracking facial changes in the perception of emotional facial expressions. The LPC factor for negative expressions correlated with the memory for emotional facial expressions. The links revealed between the latency and strength of activations of brain systems and individual differences in processing socio-emotional information provide new insights into the brain mechanisms involved in social communication.     

Holistic face representation is highly orientation-specific

It has long been argued that face processing requires disproportionate reliance on holistic processing (HP), relative to that required for nonface object recognition. Nevertheless, whether the holistic nature of face perception is achieved via a unique internal representation or by the employment of an automated attention mechanism is still debated. Previous studies had used the face inversion effect (FIE), a unique face-processing marker, or the face composite task, a gold standard paradigm measuring holistic processing, to examine the validity of these two different hypotheses, with some studies combining the two paradigms. However, the results of such studies remain inconclusive, particularly pertaining to the issue of the two proposed HP mechanisms—an internal representation as opposed to an automated attention mechanism. Here, using the complete composite paradigm design, we aimed to examine whether face rotation yields a nonlinear or a linear drop in HP, thus supporting an account that face processing is based either on an orientation-dependent internal representation or on automated attention. Our results reveal that even a relatively small perturbation in face orientation (30 deg away from upright) already causes a sharp decline in HP. These findings support the face internal representation hypothesis and the notion that the holistic processing of faces is highly orientation-specific.     

Incidental visual memory for objects in scenes

Three experiments were conducted to investigate the existence of incidentally acquired, long-term, detailed visual memory for objects embedded in previously viewed scenes. Participants performed intentional memorization and incidental visual search learning tasks while viewing photographs of real-world scenes. A visual memory test for previously viewed objects from these scenes then followed. Participants were not aware that they would be tested on the scenes following incidental learning in the visual search task. In two types of memory tests for visually specific object information (token discrimination and mirror-image discrimination), performance following both the memorization and visual search conditions was reliably above chance. These results indicate that recent demonstrations of good visual memory during scene viewing are not due to intentional scene memorization. Instead, long-term visual representations are incidentally generated as a natural product of scene perception.     

Social anxiety and attentional biases: A top-down contribution?

Selective attention toward threatening facial expressions has been found to precipitate and maintain symptoms of social anxiety. However, the automaticity of this bias is under debate. In the present study, we aimed to test whether top-down (controlled) engagement and disengagement of attention toward threatening faces is associated with social anxiety. This was examined by testing the impact of a secondary working memory (WM) load on attentional biases. In a variation of the dot-probe task, participants’ attention was initially cued to the left or right of fixation before an upright face paired with an inverted face was presented (displaying a disgust or neutral expression), and participants responded to a subsequently presented probe. The task was performed under no-load, low-load (one-digit memory task), and high-load (six-digit memory task) conditions. Social anxiety was not found to be associated with delayed disengagement from threat. However, surprisingly, high social anxiety was associated with an engagement bias away from threat, whereas low social anxiety was associated with a bias toward threat. These results were unaffected by the WM load manipulation. This indicates that engagement with threatening facial expressions has minimal contributions from top-down mechanisms, since it is likely that orienting to facial expressions occurs relatively automatically.     

No evidence for unethical amnesia for imagined actions: A failed replication and extension

In a recent study, Kouchaki and Gino (2016) suggest that memory for unethical actions is impaired, regardless of whether such actions are real or imagined. However, as we argue in the current study, their claim that people develop “unethical amnesia” confuses two distinct and dissociable memory deficits: one affecting the phenomenology of remembering and another affecting memory accuracy. To further investigate whether unethical amnesia affects memory accuracy, we conducted three studies exploring unethical amnesia for imagined ethical violations. The first study (N = 228) attempts to directly replicate the only study from Kouchaki and Gino (2016) that includes a measure of memory accuracy. The second study (N = 232) attempts again to replicate these accuracy effects from Kouchaki and Gino (2016), while including several additional variables meant to potentially help in finding the effect. The third study (N = 228) is an attempted conceptual replication using the same paradigm as Kouchaki and Gino (2016), but with a new vignette describing a different moral violation. We did not find an unethical amnesia effect involving memory accuracy in any of our three studies. These results cast doubt upon the claim that memory accuracy is impaired for imagined unethical actions. Suggestions for further ways to study memory for moral and immoral actions are discussed.     

Reconstructing the recent visual past: Hierarchical knowledge-based effects in visual working memory

This paper presents two experiments that examine the influence of multiple levels of knowledge on visual working memory (VWM). Experiment 1 focused on memory for faces. Faces were selected from continua that were constructed by morphing two face photographs in 100 steps; half of the continua morphed a famous face into an unfamiliar one, while the other half used two unfamiliar faces. Participants studied six sequentially presented faces each from a different continuum, and at test they had to locate one of these within its continuum. Experiment 2 examined immediate memory for object sizes. On each trial, six images were shown; these were either all vegetables or all random shapes. Immediately after each list, one item was presented again, in a new random size, and participants reproduced its studied size. Results suggested that two levels of knowledge influenced VWM. First, there was an overall central-tendency bias whereby items were remembered as being closer to the overall average or central tokens (averaged across items and trials) than they actually were. Second, when object knowledge was available for the to-be-remembered items (i.e., famous face or typical size of a vegetable) a further bias was introduced in responses. The results extend the findings of Hemmer and Steyvers (Psychonomic Bulletin & Review, 16, 80–87, 2009a) from episodic memory to VWM and contribute to the growing literature which illustrates the complexity and flexibility of the representations subtending VWM performance (e.g., Bae, Olkkonen, Allred, & Flombaum, Journal of Experimental Psychology: General, 144(4):744–63, 2015).     

Stuck on semantics: Processing of irrelevant object-scene inconsistencies modulates ongoing gaze behavior

People have an amazing ability to identify objects and scenes with only a glimpse. How automatic is this scene and object identification? Are scene and object semantics—let alone their semantic congruity—processed to a degree that modulates ongoing gaze behavior even if they are irrelevant to the task at hand? Objects that do not fit the semantics of the scene (e.g., a toothbrush in an office) are typically fixated longer and more often than objects that are congruent with the scene context. In this study, we overlaid a letter T onto photographs of indoor scenes and instructed participants to search for it. Some of these background images contained scene-incongruent objects. Despite their lack of relevance to the search, we found that participants spent more time in total looking at semantically incongruent compared to congruent objects in the same position of the scene. Subsequent tests of explicit and implicit memory showed that participants did not remember many of the inconsistent objects and no more of the consistent objects. We argue that when we view natural environments, scene and object relationships are processed obligatorily, such that irrelevant semantic mismatches between scene and object identity can modulate ongoing eye-movement behavior.     

Goal-relevant situations facilitate memory of neutral faces

Emotional situations are typically better remembered than neutral situations, but the psychological conditions and brain mechanisms underlying this effect remain debated. Stimulus valence and affective arousal have been suggested to explain the major role of emotional stimuli in memory facilitation. However, neither valence nor arousal are sufficient affective dimensions to explain the effect of memory facilitation. Several studies showed that negative and positive details are better remembered than neutral details. However, other studies showed that neutral information encoded and coupled with arousal did not result in a memory advantage compared with neutral information not coupled with arousal. Therefore, we suggest that the fundamental affective dimension responsible for memory facilitation is goal relevance. To test this hypothesis at behavioral and neural levels, we conducted a functional magnetic resonance imaging study and used neutral faces embedded in goal-relevant or goal-irrelevant daily life situations. At the behavioral level, we found that neutral faces encountered in goal-relevant situations were better remembered than those encountered in goal-irrelevant situations. To explain this effect, we studied neural activations involved in goal-relevant processing at encoding and in subsequent neutral face recognition. At encoding, activation of emotional brain regions (anterior cingulate, ventral striatum, ventral tegmental area, and substantia nigra) was greater for processing of goal-relevant situations than for processing of goal-irrelevant situations. At the recognition phase, despite the presentation of neutral faces, brain activation involved in social processing (superior temporal sulcus) to successfully remember identities was greater for previously encountered faces in goal-relevant than in goal-irrelevant situations.     

Systems Factorial Technology provides new insights on the other-race effect

The other-race effect refers to the difficulty of discriminating between faces from ethnic and racial groups other than one’s own. This effect may be caused by a slow, feature-by-feature, analytic process, whereas the discrimination of own-race faces occurs faster and more holistically. However, this distinction has received inconsistent support. To provide a critical test, we employed Systems Factorial Technology (Townsend & Nozawa in Journal of Mathematical Psychology, 39, 321–359, 1995), which is a powerful tool for analyzing the organization of mental networks underlying perceptual processes. We compared Taiwanese participants’ face discriminations of both own-race (Taiwanese woman) and other-race (Caucasian woman) faces according to the faces’ nose-to-mouth separation and eye-to-eye separation. We found evidence for weak holistic processing (parallel processing) coupled with the strong analytic property of a self-terminating stopping rule for own-race faces, in contrast to strong analytic processing (serial self-terminating processing) for other-race faces, supporting the holistic/analytic hypothesis.     

Cognitive-behavioral and electrophysiological evidence of the affective consequences of ignoring stimulus representations in working memory

Ignoring visual stimuli in the external environment leads to decreased liking of those items, a phenomenon attributed to the affective consequences of attentional inhibition. Here we investigated the generality of this “distractor devaluation” phenomenon by asking whether ignoring stimuli represented internally within visual working memory has the same affective consequences. In two experiments we presented participants with two or three visual stimuli and then, after the stimuli were no longer visible, provided an attentional cue indicating which item in memory was the target they would have to later recall, and which were task-irrelevant distractors. Participants subsequently judged how much they liked these stimuli. Previously-ignored distractors were consistently rated less favorably than targets, replicating prior findings of distractor devaluation. To gain converging evidence, in Experiment 2, we also examined the electrophysiological processes associated with devaluation by measuring individual differences in attention (N2pc) and working memory (CDA) event-related potentials following the attention cue. Larger amplitude of an N2pc-like component was associated with greater devaluation, suggesting that individuals displaying more effective selection of memory targets—an act aided by distractor inhibition—displayed greater levels of distractor devaluation. Individuals showing a larger post-cue CDA amplitude (but not pre-cue CDA amplitude) also showed greater distractor devaluation, supporting prior evidence that visual working-memory resources have a functional role in effecting devaluation. Together, these findings demonstrate that ignoring working-memory representations has affective consequences, and adds to the growing evidence that the contribution of selective-attention mechanisms to a wide range of human thoughts and behaviors leads to devaluation.     

Same-gender distractors are not so easy to reject: ERP evidence of gender categorization

Social categorization appears to be an automatic process that occurs during person perception. Understanding social categorization better is important because mere categorization can lead to stereotype activation and, in turn, to discrimination. In the present study we used a novel approach to examine event-related potentials (ERPs) of gender categorization in the “Who said what?” memory paradigm, thus allowing for a more in-depth understanding of the specific mechanisms underlying identity versus categorization processing. After observing video clips showing a “discussion” among female and male targets, participants were shown individual statements, each accompanied by one of the discussants’ faces. While we measured ERPs, participants had to decide whether or not a given statement had previously been made by the person with the accompanying face. In same-person trials, statements were paired with the correct person, whereas in the distractor trials, either a same-gender or a different-gender distractor was shown. As expected, participants were able to reject different-gender distractors faster than same-gender distractors, and they were more likely to falsely choose yes for a same-gender than for a different-gender distractor. Both findings indicate gender-based categorization. ERPs, analyzed in a 300- to 400-ms time window at occipito-temporal channels, indicated more negative amplitudes for yes responses both for the same person and for same-gender distractors, relative to different-gender distractors. Overall, these results show gender-based categorization even when the task was to assess the identifying information in a gender-neutral context. These findings are interpreted as showing that gender categorization occurs automatically during person perception, but later than race- or age-based categorization.