Putting category learning in order: Category structure and temporal arrangement affect the benefit of interleaved over blocked study

Recent research in inductive category learning has demonstrated that interleaved study of category exemplars results in better performance than does studying each category in separate blocks. However, the questions of how the category structure influences this advantage and how simultaneous presentation interacts with the advantage are open issues. In this article, we present three experiments. The first experiment indicates that the advantage of interleaved over blocked study is modulated by the structure of the categories being studied. More specifically, interleaved study results in better generalization for categories with high within- and between-category similarity, whereas blocked presentation results in better generalization for categories with low within- and between-category similarity. In Experiment 2, we present evidence that when presented simultaneously, between-category comparisons (interleaved presentation) result in a performance advantage for high-similarity categories, but no differences were found for low-similarity categories. In Experiment 3, we directly compared simultaneous and successive presentation of low-similarity categories. We again found an overall benefit for blocked study with these categories. Overall, these results are consistent with the proposal that interleaving emphasizes differences between categories, whereas blocking emphasizes the discovery of commonalities among objects within the same category.     

The effects of interleaving versus blocking on foreign language pronunciation learning

Many studies have shown that students learn better when they are given repeated exposures to different concepts in a way that is shuffled or interleaved, rather than blocked (e.g., Rohrer Educational Psychology Review, 24, 355–367, 2012). The present study explored the effects of interleaving versus blocking on learning French pronunciations. Native English speakers learned several French words that conformed to specific pronunciation rules (e.g., the long “o” sound formed by the letter combination “eau,” as in bateau), and these rules were presented either in blocked fashion (bateau, carreau, fardeau . . . mouton, genou, verrou . . . tandis, verglas, admis) or in interleaved fashion (bateau, mouton, tandis, carreau, genou, verglas . . .). Blocking versus interleaving was manipulated within subjects (Experiments 1–3) or between subjects (Experiment 4), and participants’ pronunciation proficiency was later tested through multiple-choice tests (Experiments 1, 2, and 4) or a recall test (Experiment 3). In all experiments, blocking benefited the learning of pronunciations more than did interleaving, and this was true whether participants learned only 4 words per rule (Experiments 1–3) or 15 words per rule (Experiment 4). Theoretical implications of these findings are discussed.     

Guidance of attention by information held in working memory

Information held in working memory (WM) can guide attention during visual search. The authors of recent studies have interpreted the effect of holding verbal labels in WM as guidance of visual attention by semantic information. In a series of experiments, we tested how attention is influenced by visual features versus category-level information about complex objects held in WM. Participants either memorized an object’s image or its category. While holding this information in memory, they searched for a target in a four-object search display. On exact-match trials, the memorized item reappeared as a distractor in the search display. On category-match trials, another exemplar of the memorized item appeared as a distractor. On neutral trials, none of the distractors were related to the memorized object. We found attentional guidance in visual search on both exact-match and category-match trials in Experiment 1, in which the exemplars were visually similar. When we controlled for visual similarity among the exemplars by using four possible exemplars (Exp. 2) or by using two exemplars rated as being visually dissimilar (Exp. 3), we found attentional guidance only on exact-match trials when participants memorized the object’s image. The same pattern of results held when the target was invariant (Exps. 2–3) and when the target was defined semantically and varied in visual features (Exp. 4). The findings of these experiments suggest that attentional guidance by WM requires active visual information.     

Why interleaving enhances inductive learning: The roles of discrimination and retrieval

Kornell and Bjork (Psychological Science 19:585–592, 2008) found that interleaving exemplars of different categories enhanced inductive learning of the concepts based on those exemplars. They hypothesized that the benefit of mixing exemplars from different categories is that doing so highlights differences between the categories. Kang and Pashler (Applied Cognitive Psychology 26:97–103, 2012) obtained results consistent with this discriminative-contrast hypothesis: Interleaving enhanced inductive learning, but temporal spacing, which does not highlight category differences, did not. We further tested the discriminative-contrast hypothesis by examining the effects of interleaving and spacing, as well as their combined effects. In three experiments, using photographs of butterflies and birds as the stimuli, temporal spacing was harmful when it interrupted the juxtaposition of interleaved categories, even when total spacing was held constant, supporting the discriminative-contrast hypothesis. Temporal spacing also had value, however, when it did not interrupt discrimination processing.     

Investigating the speed–accuracy trade-off: Better use deadlines or response signals?

Deadlines (DLs) and response signals (RSs) are two well-established techniques for investigating speed–accuracy trade-offs (SATs). Methodological differences imply, however, that corresponding data do not necessarily reflect equivalent processes. Specifically, the DL procedure grants knowledge about trial-specific time demands and requires responses before a prespecified period has elapsed. In contrast, RS intervals often vary unpredictably between trials, and responses must be given after an explicit signal. Here, we investigated the effects of these differences in a flanker task. While all conditions yielded robust SAT functions, a right-shift of the curves pointed to reduced performance in RS conditions (Experiment 1, blocked; Experiments 2 and 3, randomized), as compared with DL conditions (Experiments 1–3, blocked), indicating that the detection of the RS imposes additional task demands. Moreover, the flanker effect vanished at long intervals in RS settings, suggesting that stimulus-related effects are absorbed in a slack when decisions are completed prior to the signal. In turn, effects of a flat (Experiment 2) versus a performance-contingent payment (Experiment 3) indicated that susceptibility to response strategies is higher in the DL than in the RS method. Finally, the RS procedure led to a broad range of slow responses and high accuracies, whereas DL conditions resulted in smaller variations in the upper data range (Experiments 1 and 2); with performance-contingent payment (Experiment 3), though, data ranges became similar. Together, the results uncover characteristic procedure-related effects and should help in selection of the appropriate technique.     

Coarse guidance by numerosity in visual search

In five experiments, we examined whether the number of items can guide visual focal attention. Observers searched for the target area with the largest (or smallest) number of dots (squares in Experiment 4 and “checkerboards” in Experiment 5) among distractor areas with a smaller (or larger) number of dots. Results of Experiments 1 and 2 show that search efficiency is determined by target to distractor dot ratios. In searches where target items contained more dots than did distractor items, ratios over 1.5:1 yielded efficient search. Searches for targets where target items contained fewer dots than distractor items were harder. Here, ratios needed to be lower than 1:2 to yield efficient search. When the areas of the dots and of the squares containing them were fixed, as they were in Experiments 1 and 2, dot density and total dot area increased as dot number increased. Experiment 3 removed the density and area cues by allowing dot size and total dot area to vary. This produced a marked decline in search performance. Efficient search now required ratios of above 3:1 or below 1:3. By using more realistic and isoluminant stimuli, Experiments 4 and 5 show that guidance by numerosity is fragile. As is found with other features that guide focal attention (e.g., color, orientation, size), the numerosity differences that are able to guide attention by bottom-up signals are much coarser than the differences that can be detected in attended stimuli.     

Boundaries of semantic distraction: Dominance and lexicality act at retrieval

Three experiments investigated memory for semantic information with the goal of determining boundary conditions for the manifestation of semantic auditory distraction. Irrelevant speech disrupted the free recall of semantic category- exemplars to an equal degree regardless of whether the speech coincided with presentation or test phases of the task (Experiment 1), and this occurred regardless of whether it comprised random words or coherent sentences (Experiment 2). The effects of background speech were greater when the irrelevant speech was semantically related to the to-be-remembered material, but only when the irrelevant words were high in output dominance (Experiment 3). The implications of these findings in relation to the processing of task material and the processing of background speech are discussed.     

Are forward and backward recall the same? A dual-task study of digit recall

There is some debate surrounding the cognitive resources underlying backward digit recall. Some researchers consider it to differ from forward digit recall due to the involvement of executive control, while others suggest that backward recall involves visuospatial resources. Five experiments therefore investigated the role of executive-attentional and visuospatial resources in both forward and backward digit recall. In the first, participants completed visuospatial 0-back and 2-back tasks during the encoding of information to be remembered. The concurrent tasks did not differentially disrupt performance on backward digit recall, relative to forward digit recall. Experiment 2 shifted concurrent load to the recall phase instead and, in this case, revealed a larger effect of both tasks on backward recall, relative to forwards recall, suggesting that backward recall may draw on additional resources during the recall phase and that these resources are visuospatial in nature. Experiments 3 and 4 then further investigated the role of visual processes in forward and backward recall using dynamic visual noise (DVN). In Experiment 3, DVN was presented during encoding of information to be remembered and had no effect upon performance. However, in Experiment 4, it was presented during the recall phase, and the results provided evidence of a role for visual imagery in backward digit recall. These results were replicated in Experiment 5, in which the same list length was used for forward and backward recall tasks. The findings are discussed in terms of both theoretical and practical implications.     

Retrieval speeds context fluctuation: Why semantic generation enhances later learning but hinders prior learning

In recent work, retrieval has been shown to enhance memory for events following that retrieval. In this set of experiments, we examined the effects of interleaved semantic retrieval on both previous and future learning within a multilist learning paradigm. Interleaved retrieval led to enhanced memory for lists learned following retrieval. In contrast, memory was impaired for lists learned prior to retrieval (Experiment 1). These results are consistent with recent work in multilist learning, directed forgetting, and list-before-last retrieval, all of which indicate a crucial role for retrieval in enhancing mental list segregation. This pattern of results follows clearly from a theoretical perspective in which retrieval drives internal contextual change and in which contextual overlap between study and test promotes better memory. Consistent with that perspective, a 15-min delay before the final test eliminated both effects (Experiment 2). Experiment 2 replicated the results of Experiment 1 with materials and assessments more appropriate for educational settings: Interleaved semantic retrieval led learners to be more able to answer questions correctly about texts studied after a retrieval event but less able to do so for texts studied earlier.     

Interobject spacing explains the attentional bias toward interacting objects

Spatio-temporal interactions between simple geometrical shapes typically elicit strong impressions of intentionality. Recent research has started to explore the link between attentional processes and the detection of interacting objects. Here, we asked whether visual attention is biased toward such interactions. We investigated probe discrimination performance in algorithmically generated animations that involved two chasing objects and two randomly moving objects. In Experiment 1, we observed a pronounced attention capture effect for chasing objects. Because reduced interobject spacing is an inherent feature of interacting objects, in Experiment 2 we designed randomly moving objects that were matched to the chasing objects with respect to interobject spacing at probe onset. In this experiment, the capture effect attenuated completely. Therefore, we argue that reduced interobject spacing reflects an efficient cue to guide visual attention toward objects that interact intentionally.     

Age-of-acquisition effects in delayed picture-naming tasks

We report two experiments that explored the linguistic locus of age-of-acquisition effects in picture naming by using a delayed naming task that involved only a low proportion of trials (25 %) while, for the large majority of the trials (75 %), participants performed another task—that is, the prevalent task. The prevalent tasks were semantic categorization in Experiment 1a and grammatical-gender decision in Experiments 1b and 2. In Experiment 1a, in which participants were biased to retrieve semantic information in order to perform the semantic categorization task, delayed naming times were affected by age of acquisition, reflecting a postsemantic locus of the effect. In Experiments 1b and 2, in which participants were biased to retrieve lexical information in order to perform the grammatical gender decision task, there was also an age-of-acquisition effect. These results suggest that part of the age-of-acquisition effect in picture naming occurs at the level at which the phonological properties of words are retrieved.     

Negative emotional photographs are identified more slowly than positive photographs

In three experiments, we investigated whether the emotional valence of a photograph influenced the amount of time required to initially identify the contents of the image. In Experiment 1, participants saw a slideshow consisting of positive, neutral, and negative photographs that were balanced for arousal. During the slideshow, presentation time was substantially limited (60?ms), and the images were followed by masks. Immediately following the slideshows, participants were given a recognition memory test. Memory performance was best for positive images and worst for negative images (Experiment 1). In Experiment 2, two simultaneous photographs were briefly presented and masked. On a trial-by-trial basis, participants indicated whether the two images were identical or not, thus removing the need for memory storage and retrieval. Again, performance was worst for negative images. The results of Experiment 3 suggested that these valence-based differences were not related attentional effects (Experiment 3). We argue that the valence of an image is detected rapidly and, in the case of negative images, interferes with processing the identity of the scene.     

Top-down constraint on recognition memory

Can recognition memory be constrained “at the front end,” such that people are more likely to retrieve information about studying a recognition-test probe from a specified target source than they are to retrieve such information about a probe from a nontarget source? We adapted a procedure developed by Jacoby, Shimizu, Daniels, and Rhodes (Psychonomic Bulletin & Review 12:852–857, 2005) to address this question. Experiment 1 yielded evidence of source-constrained retrieval, but that pattern was not significant in Experiments 2, 3, and 4 (nor in several unpublished pilot experiments). In Experiment 5, in which items from the two studied sources were perceptibly different, a pattern consistent with front-end constraint of recognition emerged, but this constraint was likely exercised via visual attention rather than memory. Experiment 6 replicated both the absence of a significant constrained-retrieval pattern when the sources did not differ perceptibly (as in Exps. 2, 3 and 4) and the presence of that pattern when they did differ perceptibly (as in Exp. 5). Our results suggest that people can easily constrain recognition when items from the to-be-recognized source differ perceptibly from items from other sources (presumably via visual attention), but that it is difficult to constrain retrieval solely on the basis of source memory.     

The production effect in paired-associate learning: Benefits for item and associative information

In five experiments, we extended the production effect—better memory for items said aloud than for items read silently—to paired-associate learning, the goal being to explore whether production enhances associative information in addition to enhancing item information. In Experiments 1 and 2, we used a semantic-relatedness task in addition to the production manipulation and found no evidence of a production effect, whether the measure was cued recall or item recognition. Experiment 3 showed that the semantic-relatedness task had overshadowed the production effect; the effect was present when the semantic-relatedness task was removed, again whether cued recall or item recognition was the measure. Experiments 4 and 5 provided further evidence that production can enhance recall for word pairs and, using an associate recognition test with intact versus rearranged pairs, indicated that production may also enhance associative information. That production boosts memory for both types of information is considered in terms of distinctive encoding.     

A severe capacity limit in the consolidation of orientation information into visual short-term memory

Previous research has suggested that two color patches can be consolidated into visual short-term memory (VSTM) via an unlimited parallel process. Here we examined whether the same unlimited-capacity parallel process occurs for two oriented grating patches. Participants viewed two gratings that were presented briefly and masked. In blocks of trials, the gratings were presented either simultaneously or sequentially. In Experiments 1 and 2, the presentation of the stimuli was followed by a location cue that indicated the grating on which to base one’s response. In Experiment 1, participants responded whether the target grating was oriented clockwise or counterclockwise with respect to vertical. In Experiment 2, participants indicated whether the target grating was oriented along one of the cardinal directions (vertical or horizontal) or was obliquely oriented. Finally, in Experiment 3, the location cue was replaced with a third grating that appeared at fixation, and participants indicated whether either of the two test gratings matched this probe. Despite the fact that these responses required fairly coarse coding of the orientation information, across all methods of responding we found superior performance for sequential over simultaneous presentations. These findings suggest that the consolidation of oriented gratings into VSTM is severely limited in capacity and differs from the consolidation of color information.     

Infants’ discrimination of crossed and uncrossed horizontal disparity

In a series of preferential-looking experiments, infants 5 to 6 months of age were tested for their responsiveness to crossed and uncrossed horizontal disparity. In Experiments 1 and 2, infants were presented with dynamic random dot stereograms displaying a square target defined by either a 0.5° crossed or a 0.5° uncrossed horizontal disparity and a square control target defined by a 0.5° vertical disparity. In Experiment 3, infants were presented with the crossed and the uncrossed horizontal disparity targets used in Experiments 1 and 2. According to the results, the participants looked more often at the crossed (Experiment 1), as well as the uncrossed (Experiment 2), horizontal disparity targets than at the vertical disparity target. These results suggest that the infants were sensitive to both crossed and uncrossed horizontal disparity information. Moreover, the participants exhibited a natural visual preference for the crossed over the uncrossed horizontal disparity (Experiment 3). Since prior research established natural looking and reaching preferences for the (apparently) nearer of two objects, this finding is consistent with the hypothesis that the infants were able to extract the depth relations specified by crossed (near) and uncrossed (far) horizontal disparity.     

Exploiting illusory effects to disclose similarities in numerical and luminance processing

Recent studies have suggested that numerical and physical magnitudes are similarly processed by a generalized magnitude system. The present study investigates the number?Cluminance interaction, taking advantage of illusory effects in a cued line bisection task with numerical or nonnumerical flankers and varying levels of luminance. The results showed that both dimensions influenced bisection performance. Whereas numbers (Experiment 1) induced a systematic shift of the subjective midpoint toward the larger digit, luminance (Experiment 2) modulated the bisection performance toward the darker flanker. By combining these two illusions (Experiments 3 and 4), the two dimensions interfered with each other. This pattern of results suggests overlapping representations for physical and numerical magnitudes and highlights the value of illusory effects in cognitive research.     

Between-list lag effects in recall depend on retention interval

Although the benefits of spaced retrieval for long-term retention are well established, the majority of this work has involved spacing over relatively short intervals (on the order of seconds or minutes). In the present experiments, we evaluated the effectiveness of spaced retrieval across relatively short intervals (within a single session), as compared to longer intervals (between sessions spaced a day apart), for long-term retention (i.e., one day or one week). Across a series of seven experiments, participants (N = 536) learned paired associates to a criterion of 70 % accuracy and then received one test–feedback trial for each item. The test–feedback trial occurred within 10 min of reaching criterion (short lag) or one day later (long lag). Then, a final test occurred one day (Exps. 1–3) or one week (Exps. 4 and 5) after the test–feedback trial. Across the different materials and methods in Experiments 1–3, we found little benefit for the long-lag relative to the short-lag schedule in final recall performance—that is, no lag effect—but large effects on the retention of information from the test–feedback to the final test phase. The results from the experiments with the one-week retention interval (Exps. 4 and 5) indicated a benefit of the long-lag schedule on final recall performance (a lag effect), as well as on retention. This research shows that even when the benefits of lag are eliminated at a (relatively long) one-day retention interval, the lag effect reemerges after a one-week retention interval. The results are interpreted within an extension of the bifurcation model to the spacing effect.     

Dual-learning systems during speech category learning

Dual-system models of visual category learning posit the existence of an explicit, hypothesis-testing reflective system, as well as an implicit, procedural-based reflexive system. The reflective and reflexive learning systems are competitive and neurally dissociable. Relatively little is known about the role of these domain-general learning systems in speech category learning. Given the multidimensional, redundant, and variable nature of acoustic cues in speech categories, our working hypothesis is that speech categories are learned reflexively. To this end, we examined the relative contribution of these learning systems to speech learning in adults. Native English speakers learned to categorize Mandarin tone categories over 480 trials. The training protocol involved trial-by-trial feedback and multiple talkers. Experiments 1 and 2 examined the effect of manipulating the timing (immediate vs. delayed) and information content (full vs. minimal) of feedback. Dual-system models of visual category learning predict that delayed feedback and providing rich, informational feedback enhance reflective learning, while immediate and minimally informative feedback enhance reflexive learning. Across the two experiments, our results show that feedback manipulations that targeted reflexive learning enhanced category learning success. In Experiment 3, we examined the role of trial-to-trial talker information (mixed vs. blocked presentation) on speech category learning success. We hypothesized that the mixed condition would enhance reflexive learning by not allowing an association between talker-related acoustic cues and speech categories. Our results show that the mixed talker condition led to relatively greater accuracies. Our experiments demonstrate that speech categories are optimally learned by training methods that target the reflexive learning system.     

Face age and sex modulate the other-race effect in face recognition

Faces convey a variety of socially relevant cues that have been shown to affect recognition, such as age, sex, and race, but few studies have examined the interactive effect of these cues. White participants of two distinct age groups were presented with faces that differed in race, age, and sex in a face recognition paradigm. Replicating the other-race effect, young participants recognized young own-race faces better than young other-race faces. However, recognition performance did not differ across old faces of different races (Experiments 1, 2A). In addition, participants showed an other-age effect, recognizing White young faces better than White old faces. Sex affected recognition performance only when age was not varied (Experiment 2B). Overall, older participants showed a similar recognition pattern (Experiment 3) as young participants, displaying an other-race effect for young, but not old, faces. However, they recognized young and old White faces on a similar level. These findings indicate that face cues interact to affect recognition performance such that age and sex information reliably modulate the effect of race cues. These results extend accounts of face recognition that explain recognition biases (such as the other-race effect) as a function of dichotomous ingroup/outgroup categorization, in that outgroup characteristics are not simply additive but interactively determine recognition performance.