Perceptual integrality of sex and identity of faces: further evidence for the single-route hypothesis

According to current face-recognition models, sex (gender) and identity of faces are processed in independent routes. Using Garner's speeded-classification task, the authors provide evidence that sex and identity are processed within a single route. In 4 experiments, participants judged the sex or the familiarity of faces while the other dimension remained constant or varied randomly. The results of Experiments 1, 2, and 4 showed that participants could not selectively attend to either sex or familiarity without being influenced by the other, irrelevant dimension. Thus, identity and sex are integral dimensions. Experiment 3 provided evidence that when sex judgments are based on hairstyle heuristics, false separability can emerge. The findings support the claim that identity and sex are processed within a single route.     

The dual-task practice advantage: Empirical evidence and cognitive mechanisms

Practice of two simultaneous component tasks in dual-task situations leads to an improvement in dual-task performance. The present paper reviews empirical evidence for this practice-related improvement and discusses its underlying cognitive mechanisms. In particular, the robustness of the phenomenon of dual-task practice advantage (DTPA) is evaluated. This phenomenon is described as an advantage in dual-task performance at the end of dual-task practice (i.e., simultaneous task practice) in comparison to the dual-task performance after single-task practice (i.e., separate task practice). The review investigates the existence of empirical evidence of the DTPA phenomenon and specifications of its mechanisms in the context of the allocation and scheduling hypothesis and the integration hypothesis. Evidence for the existence of the DTPA phenomenon and the allocation and scheduling hypothesis was demonstrated in dual tasks with (1) combinations of two sensorimotor tasks, (2) combinations of one continuous task and one sensorimotor task, (3) combinations of two working-memory tasks, and (4) combinations of one motor task and one cognitive task. In contrast, the findings demonstrate that the DTPA phenomenon in dual tasks with (5) combinations of two long-term memory retrieval tasks can instead be explained by the integration hypothesis. The relevance of these findings is discussed in relation to real-world problems.

     

Processing of irrelevant location information under dual-task conditions

This study deals with the problem of whether the processing of irrelevant location information in Simon-like tasks is triggered exogenously or endogenously. In Experiment 1, the primary task required one to press, as fast as possible, a left-hand-side key or a right-hand-side key (R1) to the pitch of a tone that was presented binaurally (S1). The secondary task required identifying, without time constraints, a visual stimulus (S2) that appeared randomly to the left or right of screen center. Results showed that there was a correspondence (i.e., a cross-task Simon effect) between the location of R1 and the location of S2 when S2 was presented alone. The cross-task Simon effect became much smaller (and in-significant) when a noise stimulus was presented contralateral to S2. Experiment 2 was similar to Experiment 1, except that S2 appeared unpredictably in only one-third of the trials. Results of Experiment 2 closely replicated those of Experiment 1: the cross-task Simon effect was much greater when S2 was presented alone. Experiment 3 differed from Experiment 1 because S2 had to be processed in only one-third of the trials, in which its identity was to be reported. In the remaining two-thirds of the trials, participants could ignore S2. Results confirmed that the cross-task Simon effect was much greater when S2 was presented alone. In contrast, it did not matter whether S2 had to be processed or not. In conclusion, the present study supports the hypothesis that the task-irrelevant spatial code of the stimulus is formed automatically, likely through an exogenously triggered selection. The role of endogenously initiated selection, if any, is much less important.     

Exploring the mental number line: evidence from a dual-task paradigm

In a parity-judgment task smaller numbers are responded to faster with the left-hand key and vice versa for larger numbers (SNARC effect; Dehaene et al., in Journal of Experimental Psychology: General, 122, 371-396, 1993). We used the psychological refractory period paradigm involving a parity-judgment task and tone-discrimination task to address the question at which stage this effect arises. When the parity-judgment task is performed second, then we found equal SNARC effects for the short and the long SOA. According to the central bottleneck model, this indicates that the effect arises during the response-selection or execution stage. In Experiment 2 the parity-judgment task was performed first. The pattern of results indicates that the SNARC effect originates during the perceptual encoding or response-selection. Together, our results suggest that the SNARC effect originates while the response is selected.     

Latencies to name one of three stimulus dimensions: A study of probability effects and dimension integrality

Following each of 400 stimuli that varied according to three binary dimensions (location, color, and shape), subjects named the value of one particular dimension as quickly as possible. Each stimulus appeared on the “left” and “right” side of a screen (location dimension), and was a “red” or “green” triangle (color dimension) with apex oriented “up” or “down” (shape dimension). The alternatives of each dimension varied according to different probability distributions (90/10, 70/30, or 50/50), and three schedule conditions were contrived so that each distribution occurred once for each dimension. Nine groups of 10 subjects each defined the factorial of 3 (response conditions: identify location, color, or shape) by 3 (schedule conditions: 90 left/70 red/50 down, 70 left/50 red/90 down, or 50 left/90 red/70 down). Latencies to identify the shape dimension were influenced reliably by the probabilities of both response-relevant and response-irrelevant stimuli. Shape probabilities did not affect latencies to identify location of color, but these latencies were significantly influenced by both color and location probabilities. Results are discussed with reference to the locus of the stimulus probability effect and dimensional integrality.     

Increased control demand results in serial processing: evidence from dual-task performance

Increased demands on cognitive control trigger changes in processing mode. One such modulation involves a shift from parallel to serial processing. This study assessed the role of control demand in determining whether dual-task processing is performed serially or in parallel. We used two critical indices, based on the response-selection bottleneck model, to show that response selection was serial when a task switch was involved, but partly parallel when the simultaneous performance of the tasks did not involve task switching.     

Separate and shared sources of dual-task cost in stimulus identification and response selection

There is often strong interference if a second target stimulus (T2) is presented before processing of a prior target stimulus (T1) is complete. In the "Psychological Refractory Period" (PRP) paradigm, responses are speeded and interference manifests as increased response time for T2. In the "Attentional Blink" (AB) paradigm, stimuli are masked and responses unspeeded; interference manifests as reduced T2 accuracy. While different causes have usually been considered for PRP and AB phenomena, recent evidence has supported a unified account based on a single, shared restriction on concurrent processing. Here we show that a full assessment of separate and shared resource limitations requires direct comparison of hybrid PRP/AB trials with corresponding pure PRP and AB cases. Randomizing trial types in such a comparison also brings substantial benefit in addressing possible changes in task preparation or readiness. The data from two such experiments--combining speeded auditory (SA) and unspeeded visual (UV) task events--provide clear evidence for both separate and shared resource limitations. Often interference is strongest for T1 and T2 events of the same type, reflecting predominantly different limitations in SA and UV processing. With modest increases in demand, however, interference between different event types can also be made arbitrarily large, reflecting arbitrarily important shared limitations. For even such simple tasks as these, T--T2 interference reflects a combination of relatively local and relatively global sources.     

The role of crosstalk in dual-task performance: evidence from manipulating response-code overlap

The present study examined the role of crosstalk in dual-task interference using a combination of a nonspeeded visual task and an auditory-manual reaction time (RT) task. The potential for dual-task crosstalk was introduced by presenting in the visual task objects (e.g., a cup with a handle), which “afford” associated responses that were either spatially compatible or incompatible with the response in the RT task. Crucially, the degree of crosstalk was varied by instructing participants either to attend to the left–right orientation of the objects, creating explicit cross-task response-code overlap (“strong crosstalk”), or to attend to object identity (no direct overlap; “weak crosstalk”). The data indicated a relative benefit for cross-task compatible trials, which was much greater with strong crosstalk than with weak crosstalk. Crucially, however, even on compatible trials dual-task performance was substantially worse with strong crosstalk than with weak crosstalk. This overall cost of crosstalk suggests interference of response codes even on compatible dual-task trials. This work was presented in August 2007 at the XVth Conference of the European Society for Cognitive Psychology (ESCoP) in Marseille, France. The author would like to thank Peter A. Frensch and two anonymous reviewers for helpful comments, Julia Klotz and Marion Marksteiner for testing the participants, and Lynn Huestegge for helpful discussions.     

Anticipatory affect during action preparation: evidence from backward compatibility in dual-task performance

Upcoming responses in the second of two subsequently performed tasks can speed up compatible responses in the temporally preceding first task. Two experiments extend previous demonstration of such backward compatibility to affective features: responses to affective stimuli were faster in Task 1 when an affectively compatible response effect was anticipated for Task 2. This emotional backward-compatibility effect demonstrates that representations of the affective consequences of the Task 2 response were activated before the selection of a response in Task 1 was completed. This finding is problematic for the assumption of a serial stimulus-response translation stage. It also shows that the affective consequence of a response is anticipated during, and has an impact on stimulus-response translation, which implies that action planning considers codes representing and predicting the emotional consequences of actions. Implications for the control of emotional actions are discussed.     

Evidence for a response preparation bottleneck during dual-task performance: Effect of a startling acoustic stimulus on the psychological refractory period

The present study was designed to investigate the mechanism associated with dual-task interference in a psychological refractory period (PRP) paradigm. We used a simple reaction time paradigm consisting of a vocal response (R1) and key-lift task (R2) with a stimulus onset asynchrony (SOA) between 100 ms and 1500 ms. On selected trials we implemented a startling acoustic stimulus concurrent with the second stimulus to determine if we could involuntarily trigger the second response. Our results indicated that the PRP delay in the second response was present for both control and startle trials at short SOAs, suggesting the second response was not prepared in advance. These results support a response preparation bottleneck and can be explained via a neural activation model of preparation. In addition, we found that the reflexive startle activation was reduced in the dual-task condition for all SOAs, a result we attribute to prepulse inhibition associated with dual-task processing.     

Frequency effects in spoken and visual word recognition: evidence from dual-task methodologies

The authors report 3 dual-task experiments concerning the locus of frequency effects in word recognition. In all experiments, Task 1 entailed a simple perceptual choice and Task 2 involved lexical decision. In Experiment 1, an underadditive effect of word frequency arose for spoken words. Experiment 2 also showed underadditivity for visual lexical decision. It was concluded that word frequency exerts an influence prior to any dual-task bottleneck. A related finding in similar dual-task experiments is Task 2 response postponement at short stimulus onset asynchronies. This was explored in Experiment 3, and it was shown that response postponement was equivalent for both spoken and visual word recognition. These results imply that frequency-sensitive processes operate early and automatically.     

The role of input and output modality pairings in dual-task performance: evidence for content-dependent central interference

Recent debate regarding dual-task performance has focused on whether costs result from limitations in central capacity, and whether central operations can be performed in parallel. While these questions are controversial, the dominant models of dual-task performance share the assumption that central operations are generic--that is, their interactions are independent of stimulus and response modalities. To examine these issues, we conducted a series of dual-task experiments with different input and output modality pairings. One condition combined a visual-manual task with an auditory-vocal task, and the other condition reversed the input-output pairings, combining a visual-vocal task with an auditory-manual task. Input/output modality pairings proved to be a key factor; throughout practice, dual-task costs were generally more than twice as large with visual-vocal/auditory-manual tasks than with the opposite arrangement of modalities (Experiments 1 and 2). These differences could be explained neither by competition for peripheral resources nor by differences in single-task response times (Experiment 3). Moreover, the persistent dual-task costs did not appear to stem from a central bottleneck. Contrary to the dominant models of dual-task performance, we propose that central interference between tasks depends not just on the duration of central operations, nor just strategic adaptation, but also on the content of those operations. Implications for structural and strategic accounts of dual-task interference are discussed.     

A dual-task analysis of detection accuracy for the case of high target-distractor similarity: Further evidence for independent processing

Previous studies of visual detection performance indicate that perceptual span has a negligible effect on detection accuracy; hence, previous results suggest that individual characters within a multielement array are perceived independently. In this study, additional experimental factors were examined in order to determine whether conditions exist in which perceptual span influences detection performance: targets and distractors were chosen to be maximally similar, and detection accuracy was studied in relation to display size, target redundancy, and homogeneity of distractor characters. To test for perceptual span effects, a three-state model was proposed which incorporates the assumption that individual characters are perceived independently. The model may be applied to (a) both yes-no and forced-choice tasks, and (b) the case when effects of display characteristics differ for response alternatives. Predictions based on the three-state model agree satisfactorily with observed proportions. Moreover, interactions between display characteristics and response alternatives accord well with model predictions, and accuracy differences between subjects are consistent with individual differences in model parameters. The results provide further evidence for the independence assumption and illustrate additional conditions for which perceptual span has a negligible effect on detection accuracy.     

Capacity demands of phoneme selection in word production: new evidence from dual-task experiments

Three dual-task experiments investigated the capacity demands of phoneme selection in picture naming. On each trial, participants named a target picture (Task 1) and carried out a tone discrimination task (Task 2). To vary the time required for phoneme selection, the authors combined the targets with phonologically related or unrelated distractor pictures (Experiment 1) or words, which were clearly visible (Experiment 2) or masked (Experiment 3). When pictures or masked words were presented, the tone discrimination and picture naming latencies were shorter in the related condition than in the unrelated condition, which indicates that phoneme selection requires central processing capacity. However, when the distractor words were clearly visible, the facilitatory effect was confined to the picture naming latencies. This pattern arose because the visible related distractor words facilitated phoneme selection but slowed down speech monitoring processes that had to be completed before the response to the tone could be selected.     

Processing capacity requirements of stimulus encoding.

Processing capacity requirements of stimulus encoding are examined in the light of conflicting evidence from studies by Posner and Boies (1971) and Comstock (1973). Capacity demand is assessed by impairment in performance on a secondary probe detection task at the moment of encoding on the primary task. The results support Comstock's (1973) conclusion that stimulus encoding does require processing capacity and relate capacity demand to the level of encoding difficulty. Evidence suggests that it costs more capacity to share attention between two modalities rather than within one. The importance of secondary task sensitivity in assessing capacity requirements is also illustrated.     

Manipulation of stimulus onset delay in reading: evidence for parallel programming of saccades

On-line eye movement recording of 12 subjects who read short stories on a cathode ray tube enabled a test of direct control and preprogramming models of eye movements in reading. Contingent upon eye position, a mask was displayed in place of the letters in central vision after each saccade, delaying the onset of the stimulus in each eye fixation. The duration of the delay was manipulated in fixed or randomized blocks. Although the length of the delay strongly affected the duration of the fixations, there was no difference due to the conditions of delay manipulation, indicating that fixation duration is under direct control. However, not all fixations were lengthened by the period of the delay. Some ended while the mask was still present, suggesting they had been preprogrammed. But these "anticipation" eye movements could not have been completely determined before the fixation was processed because their fixation durations and saccade lengths were affected by the spatial extent of the mask, which varied randomly. Neither preprogramming nor existing serial direct control models of eye guidance can adequately account for these data. Instead, a model with direct control and parallel programming of saccades is proposed to explain the data and eye movements in reading in general.     

Perceptual learning in flavor aversion: evidence for learned changes in stimulus effectiveness

Rats were exposed to the compound flavors AX and BX, presented in alternation, and to CX on a separate block of trials. Generalization to BX after aversion conditioning with AX was less than to CX. An equivalent effect was found when the nature of the common element was changed after preexposure but not when the common element was omitted during preexposure, during conditioning and test, or both. Rats conditioned with X alone again showed less aversion to BX than to CX; similarly, rats conditioned with a novel flavor (Y) showed less aversion to BY than to CY. These effects support the proposal that intermixed preexposure to AX and BX enhances the perceptual effectiveness of their unique features, A and B.     

Processing of sequential tactile patterns: effects of a neutral stimulus

The perception of a target pattern may be interfered with by the presentation of a nontarget pattern in close temporal and spatial proximity with it. The results from previous studies suggested that much of this interference is the result not of masking but of response competition, subjects responding with the nontarget instead of the target. Using a 4-to-2 paradigm in which four target patterns are mapped onto two responses, it was shown that neutral patterns (i.e., patterns with no responses associated with them) produce considerable interference. The amount of interference is less than that produced by patterns associated with incorrect responses but greater than that produced by patterns associated with correct responses. The amount of interference produced by neutral patterns did not vary as a function of the form of the neutral pattern (Experiments 1 and 2); however, the amount of interference did depend on the degree and nature of the similarity between the neutral and target patterns (Experiments 3 and 4). The results indicate that recent studies have underestimated the amount of interference due to masking and overestimated the amount due to response competition. Response competition may either hinder or help target categorization depending on the nontarget pattern.