Eliminating flanker effects and negative priming in the flankers task: evidence for early selection

Reports of negative priming in the absence of flanker effects (Fox, 1995) provide support for the notion that unattended stimuli are identified. I evaluated the hypothesis that such results are the outcome of attentional leakage to the flanker location. In Experiment 1, I assessed flanker effects and negative priming as a function of target-flanker proximity (.9 degrees and 2.7 degrees for near and far flankers, respectively) and of attention cuing to the target location (precued vs. uncued) on the prime trials. I report larger flanker effects in uncued than in precued conditions, and larger effects for near than for far flankers. More critically, when attention was precued, both flanker effects and negative priming vanished for far flankers. In Experiment 2, I show that the latter result was not linked to prime-probe contextual similarity (Neill, 1997). These results demonstrate that selective target processing is possible when attention is optimally focused to the target location.     

Flanker and negative flanker effects in letter identification

In a speeded choice reaction time task, responses to centrally presented letter targets can be altered by the identity of surrounding task-irrelevant letters (flankers). In the standard flanker effect, flankers associated with the same response as the target lead to faster and more accurate responses, whereas flankers associated with a different response lead to slower and more error-prone responses. B. A. Eriksen and C. W. Eriksen (1974, pp. 143-149) have argued that these flanker effects occur through response competition. We present data from a novel version of the Eriksen task, in which some targets and flankers consist of letter forms that are morphed versions of target letters. In this paradigm, flankers induce classic flanker effects on well-formed targets. But flankers induce an opposite effect, termed a negative flanker effect on morphed letter targets. For example, targets that are morphs between the letters "A" and "H" are more likely to be identified as an "A" when flanked by an "H." The interpretation advanced here is that there are two distinct kinds of flanker effects contrast enhancement in perceptual processes and response competition in response selection processes.     

Attentional capture by emotional faces is contingent on attentional control settings

Attentional capture by schematic emotional faces was investigated in two experiments using the flanker task devised by Eriksen and Eriksen (1974). In Experiment 1, participants were presented with a central target (a schematic face that was either positive or negative) flanked by two identical distractors, one on either side (schematic faces that were positive, negative, or neutral). The objective was to identify the central target as quickly as possible. The impact of the flankers depended on their emotional expression. Consistent with a threat advantage hypothesis (negative faces are processed faster and attract more processing resources), responses to positive faces were slower when these were flanked by (response incompatible) negative faces as compared with positive or neutral faces, whereas responses to negative faces were unaffected by the identity of the flankers. Experiment 2 was a standard flanker task with letter stimuli except that the task-neutral flankers were schematic faces that were either positive, negative, or emotionally neutral. In this case, in which faces and emotional expressions were to be ignored, performance seemed entirely unaffected by the faces. This result suggests that attentional capture by emotional faces is contingent on attentional control settings.     

Attentional capture by emotional faces is contingent on attentional control settings

Attentional capture by schematic emotional faces was investigated in two experiments using the flanker task devised by Eriksen and Eriksen (1974). In Experiment 1, participants were presented with a central target (a schematic face that was either positive or negative) flanked by two identical distractors, one on either side (schematic faces that were positive, negative, or neutral). The objective was to identify the central target as quickly as possible. The impact of the flankers depended on their emotional expression. Consistent with a threat advantage hypothesis (negative faces are processed faster and attract more processing resources), responses to positive faces were slower when these were flanked by (response incompatible) negative faces as compared with positive or neutral faces, whereas responses to negative faces were unaffected by the identity of the flankers. Experiment 2 was a standard flanker task with letter stimuli except that the task-neutral flankers were schematic faces that were either positive, negative, or emotionally neutral. In this case, in which faces and emotional expressions were to be ignored, performance seemed entirely unaffected by the faces. This result suggests that attentional capture by emotional faces is contingent on attentional control settings.     

无意识信息的刺激表征及其时间特性

无意识信息是否存在刺激表征有待解决。实验采用字母Flanker任务, 通过目标和flanker在刺激水平和反应水平上的冲突效应, 重点考察阈下flanker在刺激水平的表征。在相同的实验设计下, flanker可觉察时(实验1A), 观察到经典的刺激冲突效应和反应冲突效应; 而flanker无意识时(实验1B), 重复了反应冲突效应, 却观察到刺激冲突效应发生反转, 提示了无意识信息的反应加工及刺激加工。实验2采用符号材料促使刺激−反应的自动联结, 减少刺激−反应规则的难度, 结果重复了实验1B的发现--无意识信息引起反转的刺激冲突。刺激冲突下反转效应的一致结果表明无意识信息的刺激表征影响了认知加工。实验3考察这种影响在时间进程上的特点。无意识刺激表征的影响随反应时变化：在快速反应中观察到刺激冲突效应, 之后该效应发生反转且反转量随反应时增加而增大。相反, 无意识反应表征的影响在不同反应时下保持稳定。以上结果提供了无意识信息存在刺激表征的行为学证据并揭示了其时间特性, 提供了无意识领域中不一致研究结果的整合思路。     

A bottleneck in face identification: repetition priming from flanker images

There is evidence that face processing is capacity-limited in distractor interference tasks and in tasks requiring overt recognition memory. We examined whether capacity limits for faces can be observed with a more sensitive measure of visual processing, by measuring repetition priming of flanker faces that were presented alongside a face or a nonface target. In Experiment 1, we found identity priming for face flankers, by measuring repetition priming across a change in image, during task-relevant nonface processing, but not during the processing of a concurrently-presented face target. Experiment 2 showed perceptual priming of the flanker faces, across identical images at prime and test, when they were presented alongside a face target. In a third Experiment, all of these effects were replicated by measuring identity priming and perceptual priming within the same task. Overall, these results imply that face processing is capacity limited, such that only a single face can be identified at one time. Merely attending to a target face appears sufficient to trigger these capacity limits, thereby extinguishing identification of a second face in the display, although our results demonstrate that the additional face remains at least subject to superficial image processing.     

The influence of divided attention on holistic face perception.

There is evidence that upright, but not inverted, faces are encoded holistically. The holistic coding of faces was examined in four experiments by manipulating the attention allocated to target faces. In Experiment 1, participants in a divided attention condition were asked to match two upright flanker faces while encoding a centrally presented upright target face. Although holistic coding was evident in the full attention conditions, dividing attention disrupted holistic coding of target faces. In Experiment 2, we found that while matching upright flanker faces disrupted holistic coding, matching inverted flanker faces did not. Experiment 3 demonstrated that the differential effects of flanker orientation were not due to participants taking longer to match upright, than inverted, flanker faces. In Experiment 4, we found that matching fractured faces had an intermediate effect to that of matching upright and inverted flankers, on the holistic coding of the target faces. The findings emphasize the differences in processing of upright, fractured and inverted faces and suggest that there are limitations in the number of faces that can be holistically coded in a brief time.     

Intentional versus unintentional use of contingencies between perceptual events

In three experiments we studied human ability to use statistical contingencies between visual stimuli (flankers and targets) to improve performance in a letter-digit classification task. We compared the performance of explicitly informed subjects with that of subjects who were told nothing of the contingencies. Simultaneous presentation of flankers and targets (Experiment 1) produced evidence of unintentional contingency use by both informed and uninformed subjects. When stimuli on trialn predicted target stimuli on trialn+1 (Experiment 2) there was no evidence of unintentional processes, but informed subjects showed strong evidence of using intentional prediction strategies. When flanker onset preceded target stimuli presentation (Experiment 3), evidence of contingency use by both informed and uninformed subjects was found, but the data illustrated qualitative differences in response style (e.g., speed-accuracy tradeoffs) between the two groups. Intentional and unintentional uses of contingencies between perceptual events are qualitatively distinct with respect to the time frame in which they can be applied and the performance patterns they produce. Finally, we argue that the unintentional processes studied here are implicit in nature.     

Correlational cuing as a function of target complexity and target-flanker similarity

It is generally assumed that the correlational cuing effect (CE) between targets and correlated flankers is due to learning association between the flankers and their correlated responses. The present study challenges this view. Experiment 1 shows that the CE for targets composed of color is eliminated as soon as the correlation is removed. Experiment 2 shows that the CE during training is not due to association of the flankers with responses. Experiment 3 shows that at least some of the CE during training with the correlation is due to repetition priming of the display. Experiment 4 replicates the results of Experiment 1 for orientation targets. In Experiments 5-7, more typical tasks with letter targets are examined, and it is demonstrated that preexperimental similarity between targets and correlated flankers is crucial. The CE for correlated but dissimilar target-flanker pairs, similar to that for color and orientation targets, is confined to on-line processes that occur during training. The CE is transferred, however, for correlated and similar target-flanker pairs. We propose that, at least for the simple stimulus to response mapping used in our study, the CE is not due to learning at all. Instead it is due to (1) on-line processes, such as repetition priming, that occur during training with the correlation and (2) a regular flanker effect (see, e.g., B. A. Eriksen & C. W. Eriksen, 1974) that occurs for similar target-flanker pairs.     

Testing the attention shift hypothesis as an account for the flanker sequence-based congruency modulation in spatial flanker tasks

Smaller Simon effects when stimulus locations are repeated on successive trials rather than alternated have been explained by the attention shift hypothesis, suggesting that shifts of attention result in interfering response codes. We investigated whether the attention shift hypothesis can also explain smaller flanker effects for repeated flankers than for alternated flankers, which occur only with directional information. In 3 peripheral letter identification tasks, target locations were cued by partial or complete flanker stimuli. Experiments 1 and 2 showed that directional flankers elicit shifts of attention. However, Experiment 3 revealed that directional flankers induced inverted cuing effects when reacting to the central target arrow was additionally required. These results are difficult to reconcile with the attention shift hypothesis as an explanation for the congruency reduction with repetitions of directional flankers.     

Flank to the left,flank to the right: Testing the modified receptive field hypothesis of letter-specific crowding

The present study tested for effects of number of flankers positioned to the left and to the right of target characters as a function of visual field and stimulus type (letters or shapes). On the basis of the modified receptive field hypothesis (Chanceaux & Grainger, 2012), we predicted that the greatest effects of flanker interference would occur for leftward flankers with letter targets in the left visual field. Target letters and simple shape stimuli were briefly presented and accompanied by either 1, 2, or 3 flankers of the same category either to the left or to the right of the target, and in all conditions with a single flanker on the opposite side. Targets were presented in the left or right visual field at a fixed eccentricity, such that targets and flankers always fell into the same visual field. Results showed greatest interference for leftward flankers associated with letter targets in the left visual field, as predicted by the modified receptive field hypothesis.     

Orientation asymmetry in the flanker task

Experiments 1 and 2 of this study show that when the target is either a vertical or a horizontal line, diagonal-line flankers tilted 45° either to the right or to the left have the same effect as do incongruent flankers. When the target is a diagonal line tilted either to the right or to the left, vertical- or horizontal-line flankers do not have the same effect as do incongruent flankers. Experiment 3 demonstrates that this asymmetry is not caused by the temporal-dynamic aspects of the processing. Together, these experiments suggest that there is an asymmetrical relation between diagonal lines and either vertical or horizontal lines otttside of the central focus of attention. Experiment 4 shows that despite this asymmetry in the flanker task, visual search for a vertical- or a horizontal-line target among diagonal-line distractors is not affected by the number of distractors. Possible explanations of this phenomenon are discussed.     

Lateral masking: limitations of the feature interaction account

Recognition performance for a target letter embedded in a string of characters is worse than that for targets presented in isolation. This lateral masking (LM) effect is known to depend on target eccentricity and spacing between target and flankers (Bouma, 1970), indicating that LM arises in early visual processing due to interactions among visual features. The feature interaction account would predict that flankers consisting of similar features produce similar LM effects and that differences in LM produced by different types of flanker diminish with increasing target eccentricity and decreasing spacing. However, in a series of six experiments, this prediction was shown not to be true. Flankers that did not access a higher level code (e.g., pseudoletters or rotated letters) produced more LM than standard letter flankers. Moreover, effects of different flanker types were most pronounced for medium target eccentricities and medium spacings for which recognition performance scores ranged between 40% and 60%.     

Attentional processing of “unattended” flankers: Evidence for a failure of selective attention

Results from research with the flanker task have been used to argue both that flankers are identified without attention and that flanker identificatin requires attention. In three experiments, we addressed this issue by examining flanker recall. In Experiment 1, we manipulated flanker redundancy, a variable that could influence attention to the flankers, in order to determine whether it affected the magnitude of the flanker effect, the magnitude of flanker recall, or both. Redundancy did not influence the flanker effect, and recall was high in both conditions, suggesting that the flankers were attended. The recall results contradicted those of Miller (1987), the only other study that we could find in which flanker recall was used. In Experiments 2 and 3, we examined differences between Miller’s procedure and ours. Although flanker recall was lower when open-ended rather than forced recall was used and when the flanker task was made more complicated, flanker recall remained well above chance in all conditions. The data strongly suggest that in the typical correlated flanker task there is a failure of selective attention at some point during processing such that flankers receive attentional processing.     

Differences between digit naming and number word reading in a flanker task

Numbers can be represented as number words or as digits, but are the two notations processed differently? Two experiments in which a flanker paradigm with a naming task was used were conducted, with digits and number words as targets and flankers. Reaction times were shortest when the flanker denoted the same numerical value as the target. The numerical distance between the target and a numerically different flanker modulated reaction times in all conditions, except for number word targets with digit flankers. The direction of this effect—targets were named faster when the flanker was numerically close than when it was far—indicates that the numerical magnitude representations of numbers are associatively connected. When the target and the flanker were presented in the same format, no difference was observed in the distance effects for the two formats. This indicates that number words activate the abstract representation of their numerical value in a way that is very similar to that for digits.     

The Eriksen flanker effect revisited.

Four studies are reported on the potential role of perceptual interference in a standard Eriksen flanker task. In the first study, incongruent flanker letters showed the usual effect on choice reaction time (CRT) to the target letter but had no effect on the visual fixation time (VFT) needed to distinguish target and flankers. In the second experiment, the effect of incongruent flankers was studied in the context of a same-different response in regard to the target letter and a subsequently presented single letter. The effect of incongruent flankers vanished at an interstimulus interval of 200 ms. In Experiment 3, the same-different task was used in the paradigm of the functional visual field with a target-flankers combination as stimulus on the left (SL) and a single letter as stimulus on the right side (SR) of the visual field. Flankers did neither affect VFT nor the same-different CRT suggesting that target selection may proceed during the saccade from SL to SR. In Experiment 4 effects were studied of flanker-to-target and target-to-single-letter similarity. Flanker-to-target similarity did neither affect VFT nor same-different CRT but target-to-single-letter similarity prolonged same-different CRT. Together, the results suggest parallel perceptual processing of target and flankers, followed by competition of responses to the target and to the incongruent flankers. In line with earlier research, processes of response selection and response competition appear not to be tied to VFT but to proceed in parallel with the saccade from SL to SR.     

Limits on the application of additive factors logic: Violations of stage robustness suggest a dual-process architecture to explain flanker effects on target processing

This study aimed to establish whether the logic of the AFM applies when multielement stimuli contain relevant and irrelevant elements. Target Size (TS) and symbolic S-R Compatibility (SRC) were manipulated in three reaction time (RT) experiments. TS and SRC are assumed in the AFM literature to selectively influence the independent stages of feature extraction and S-R translation, respectively. Experiment 1 showed that the effects of TS and SRC on RT were additive when the target was presented in isolation and this additive relation was not changed when the target was flanked by stimuli that contained no information relevant to the response. In Experiment 2, this additivity changed into a superadditive interaction when flankers signaled the same response as the target: The effect of SRC was larger when targets were small rather than large. The overall pattern of findings violated the AFM stage robustness criterion. Neither a discrete stage model nor a continuous flow conception account for the results. To explain flanker effects on target processing a dual-process architecture was formulated that assumes that perceptual information is processed along concurrently engaged routes: An attentive processing route and a direct priming route. Experiment 3 confirmed the prediction of the dual-process model that the relation between TS and SRC would be subadditive when flankers signal the response opposite to that designated by the target.     

Flanker effects on motor output and the late-level response activation hypothesis.

When participants must respond to a relevant central target and ignore irrelevant flanking stimuli, the flankers produce a flanker compatibility effect on behavioural measures. Current accounts of the flanker compatibility effect assume that both target and flanker stimuli affect response activation. This idea is supported by electrophysiological studies, which show that irrelevant flanker stimuli can affect the motor system. The present experiments examined the characteristics of flanker effects on the motor system by analysing the details of the motor output with response force measures. A total of 60 participants responded in the flanker task to arrows (Experiment 1) or letters (Experiment 2). Reaction time as well as response force increased on incompatible trials. Analyses of the distribution of incorrect activation revealed that both response times and correct motor output increased with the amount of incorrect activity. However, the flanker compatibility effect was only marginally modulated by incorrect activity. Results suggest that the largest part of the flanker compatibility effect cannot be attributed to response activation and competition at late levels of the response system.     

Effect of colour pop-out on the recognition of letters in crowding conditions

The crowding effect of adjacent objects on the recognition of a target can be reduced when target and flankers differ in some feature, that is irrelevant to the recognition task. In this study, the mechanisms of this effect were explored using targets and flankers of the same and different colours. It was found that facilitation nearly equal to that of differently coloured targets and flankers can be observed with a differently coloured background blob in the location of the target. The different-colour effect does not require advance knowledge of the target and flanker colours, but the effect increases in the course of three trials with constant mapping of colours. The results are consistent with the notion of exogenous attention that facilitates the processing at the most salient locations in the visual field.     

Strategies of flanker coprocessing in single and dual tasks

In this study, the authors used a dual-task flanker paradigm to investigate the degree to which flankers are coprocessed with the target as a function of whether flankers have to be used as stimuli for a second task. A series of experiments, in which performance in dual tasks was compared with that in single tasks, revealed that participants had a strong tendency to coprocess flankers to a large degree in dual tasks, even if this impaired performance. Coprocessing of flankers was reduced only when totally irrelevant flankers were presented at the beginning of a trial or single tasks were performed on the great majority of trials within a block. The results suggest that it was demanding to process targets and flankers serially when both had to be used for a dual task. As a consequence, target and flankers were processed in parallel, even if this was nonoptimal for target selection.