حواس پرتی توسط بستگان: اثر آسیب لوب فرونتال بر روی حواس پرتی معنایی

Abstract When young adults carry out visual search, distractors that are semantically related, rather than unrelated, to targets can disrupt target selection (see Belke et al., 2008 and Moores et al., 2003). This effect is apparent on the first eye movements in search, suggesting that attention is sometimes captured by related distractors. Here we assessed effects of semantically related distractors on search in patients with frontal-lobe lesions and compared them to the effects in age-matched controls. Compared with the controls, the patients were less likely to make a first saccade to the target and they were more likely to saccade to distractors (whether related or unrelated to the target). This suggests a deficit in a first stage of selecting a potential target for attention. In addition, the patients made more errors by responding to semantically related distractors on target-absent trials. This indicates a problem at a second stage of target verification, after items have been attended. The data suggest that frontal lobe damage disrupts both the ability to use peripheral information to guide attention, and the ability to keep separate the target of search from the related items, on occasions when related items achieve selection.

1. Introduction In visual search tasks participants are asked to decide whether a pre-specified target is present on the screen. Many theories assume that search is guided to a target by an “attentional template” held in working memory. Evidence for such a template comes from a number of sources. Chelazzi, Miller, Duncan, and Desimone (1993) trained monkeys to make a saccade to an item in a search display that matched a stimulus held in working memory (using a match to sample task). They found that cells in the inferior temporal lobe responding to the cued item maintained their activity during the interval between the cue and the search display, with the cells showing an enhanced rise in activation when the cued item re-appeared in the search display. Chelazzi et al. proposed that the activity maintained during the interval between the cue and the display represented a template that biased activity in earlier cortical regions to favor features consistent with the target. Evidence for effects of top-down guidance on human search comes from a number of sources. For example, several investigators have reported asymmetries in visual search, with search tasks varying in difficulty according to which item is the search target and which is the distractor (e.g., a large target versus small distractors generates efficient search, whereas a small target amongst large distractors generates inefficient search; Wolfe (1998)). Hodsoll and Humphreys (2001) showed that this search asymmetry was modulated by fore-knowledge of the target: the asymmetry was larger when participants knew what they were searching for relative to when they searched for a target that was the odd one out (see also Hodsoll & Humphreys (2005), for similar evidence from orientation search asymmetries). Hodsoll and Humphreys proposed that the search asymmetry was partially dependent on the match between the stimulus and the search template (some stimuli are matched more quickly than others) and not just on bottom-up differences between the stimuli. Moores et al. (2003) provided other evidence for a template by assessing the effects of semantic distractors on search. They asked participants to search for a familiar target object (e.g., motorbike) and, on some trials, presented semantic distractors in the display (e.g., motorbike helmet). They found that reaction times were slowed on trials when semantic distractors were present. On target-absent trials in particular, the first eye movement tended to go to the semantically related distractor rather than to unrelated distractors. These data suggest that activation of a memory template for a target also excites the re-presentations of related items, which can then guide search to matching (but in this case, distractor) stimuli. In the present paper, we use data from patients with frontal-lobe lesions to probe-apart the different processes involved in guiding search to targets. In particular, using the procedure of Moores et al. (2003) we examine whether, in addition to any effects of related distractors on the initial stages of selection, there are effects at later stages in which any selected stimuli are compared with target-related templates. At what stage(s) can semantic information about the stimuli be accessed to influence target selection and how is this affected by damage to the frontal lobes? We present data suggesting that although there are effects of target-distractor relatedness at both stages and across patient and control groups, frontal lobe damage alters: (i) the initial gathering of peripheral information that guides the first stages of target selection (which is independent of target-distractor relatedness) and (ii) the later process of target identification following the orienting of attention to a stimulus (which is affected by target-distractor relatedness).

6. Conclusions The data indicate that patients with frontal-lobe lesions can have problems at two stages of visual search: (i) impaired selection of targets in relation to both related and unrelated distractors, and (ii) impairments in rejecting semantically related distractors at a post-selection stage of target identification. The data fit with the proposal that the frontal lobes help to specify the target for search and to keep separate the re-presentation of task-relevant targets from activated re-presentations of other items. In addition, the AC may play a particular role in monitoring conflict when distractors prime responses that are at variance with the response required in the search task.