Elsevier

Neuropsychologia

Volume 47, Issue 6, May 2009, Pages 1585-1591
Neuropsychologia

Both egocentric and allocentric cues support spatial priming in visual search

https://doi.org/10.1016/j.neuropsychologia.2008.11.017Get rights and content

Abstract

The perception–action model proposes that vision for perception and vision for action are subserved by two separate cortical systems, the ventral and dorsal streams, respectively [Milner, A. D., & Goodale, M. A. (1995). The visual brain in action (1st ed.). Oxford: Oxford University Press; Milner, A. D., & Goodale, M. A. (2006). The visual brain in action (2nd ed.). Oxford: Oxford University Press Inc.]. The dorsal stream codes spatial information egocentrically, that is, relative to the observer. Egocentric representations are argued to be highly transient; therefore, it might be expected that egocentric information cannot be used for spatial memory tasks, even when the visual information only needs to be retained for a few seconds. Here, by applying a spatial priming paradigm to a visual search task, we investigated whether short-term spatial memory can use egocentric information. Spatial priming manifests itself in speeded detection times for a target when that target appears in the same location it previously appeared in. Target locations can be defined in either egocentric or allocentric (i.e. relative to other items in the display) frames of reference; however, it is unclear which of these are used in spatial priming, or if both are. Our results show that both allocentric and egocentric cues were used in spatial priming, and that egocentric cues were in fact more effective than allocentric cues for short-term priming. We conclude that egocentric information can persist for several seconds; a conclusion which is at odds with the assumption of the perception–action model that egocentric representations are highly transient.

Introduction

The locations of objects in a visual environment can be defined using different frames of reference. Egocentric frames of reference (observer-based metrics) define spatial positions using the body, or a specific part of the body, for instance, the trunk or the head, as a constant point of reference. Conversely, allocentric frames of reference (object-based metrics) consider spatial relations between objects, and rely on the external environment to define space, that is, by using landmarks (Burgess, Spiers, & Paleologou, 2004). Milner and Goodale (1993) argue that visuomotor control relies on representations which are centred on the individual: to execute the appropriate and correctly scaled motor action the individual needs to know where the object is in relation to them self and the effectors that will carry out the action. Furthermore, Milner and Goodale (1993) argue that owing to the fact that the position of our head, body, and eyes relative to external objects constantly change, the dorsal stream does not rely on stored representations. They use this assumption to explain the complementary effects of temporal delays on the visuomotor performance of patients with either dorsal or ventral stream damage (Goodale, Jakobson, & Keillor, 1994; Milner et al., 2001; Milner, Paulignan, Dijkerman, Michel, & Jeannerod, 1999).

Supporting the view that the dorsal stream cannot store spatial information, Westwood, Heath, and Roy (2001) found greater pointing errors when the target was occluded for 500 ms before movement was cued compared to when the movement was cued immediately, thereby suggesting very limited availability of dorsal stream representations. Additionally Westwood and Goodale (2003) observed that although visual illusions do not affect motor performance when the action is executed immediately (Aglioti, DeSouza, & Goodale, 1995), visual illusions have a significant effect on visuomotor performance already after delays of 2.5 s. This suggests that after a delay it is the ventral rather than the dorsal stream, which provides the visual information for the control of the movements. It is also suggested that there would be no advantage to storing egocentric representations after they have either been used to execute a particular movement or they are no longer relevant to the particular movement; therefore, suggesting that egocentric representations have a short time span (Westwood, Heath, & Roy, 2003). Given that the relevant information in visuomotor control is egocentric in origin, and that egocentric information has been associated with dorsal stream structures, one might therefore expect that egocentric information cannot be stored for more than 2 s, and that tasks, which depend on visual memory, have to rely on allocentric representations.

The research into topographical memory has not been conclusive about whether egocentric representations can be stored; for example, while some have shown that changing the egocentric information between the learning phase and the testing phase causes less disruption than changing the allocentric information (Burgess et al., 2004; Simons & Wang, 1998; Wang & Simons, 1999), others report that recognition performance of object location is affected by egocentric changes (Christou & Buelthoff, 1999; Diwadkar & McNamara, 1997; Finlay, Motes, & Kozhevnikov, 2007; Shelton & McNamara, 2004). However, in most of these studies participants are required to report whether the location of an object has changed relative to other objects or landmarks in the scene, thus by its very nature the task prompts participants to employ allocentric information. Furthermore, generally these studies used long delays (7–13 s). Therefore, they cannot be used to address the question of whether egocentric information might be useful for short delays of a few seconds, a question which is of obvious interest in the context of Milner and Goodale's perception and action model since they used the dorsal stream's inability to store information, even for more than 2 s, to explain the significant effects of such short delays on visuomotor performance in optic ataxia, visual form agnosia, and studies using visual illusions.

To address this question, there are three conditions which must be met in order to accurately assess the duration and use of egocentric spatial information. Firstly, spatial memory must be tested over very short time spans. Secondly, performance in an egocentric condition must be compared directly with that in an allocentric condition. To allow a direct comparison between these two conditions the same behavioural task must be used. Since in Milner and Goodale's model allocentric information is associated with perceptual tasks while egocentric information is associated with visuomotor tasks, there is always a risk that using either a perceptual or a visuomotor task might provide a bias against the use of egocentric or allocentric information, respectively. In our study we decided to use a perceptual task to provide a stronger test of the hypothesis that egocentric information is unavailable after a delay. Since a perceptual task might bias against the use of egocentric information any evidence of its use after a delay has to be taken even more seriously. Furthermore using a perceptual task also allows us to address another interesting question, namely whether Milner and Goodale's thesis that egocentric information is not relevant for typical perceptual tasks is correct. Finally, participants should not be prompted towards using one type of spatial information, so no instructions about how the target is coded should be provided.

An experimental paradigm, which fulfils these criteria, is priming in visual search. A standard visual search task requires the observer to decide whether a specified item, the target, is present in the display or not; the time taken to make this judgement is known as the search time and provides an index of search efficiency (Wolfe, 1998). Priming in visual search refers to the influence that memory processes can have from one trial to the next or within the same trial (Shore & Klein, 2001). Between-trial memory, known as trial-to-trial priming, operates over a time scale in the order of seconds to minutes. In trial-to-trial priming experience of a stimulus can influence future encounters with that same stimulus, specifically, repetition of a feature can facilitate the processing and speed of detection of that feature. It is likely that this repetition priming results from a memory representation of the first trial being stored and which is retrieved when that same trial is presented a second time, leading to more efficient processing of that stimulus. Trial-to-trial priming has been shown for target colour (Hilstrom, 2000; Huang, Holcombe, & Pashler, 2004; Maljkovic & Nakayama, 1994) and target location (Kristjansson, Vuilleumier, Malhotra, Husain, & Driver, 2005; Kumada & Humphreys, 2002; Maljkovic & Nakayama, 1996).

Although it is clear that spatial position can be primed, it is not clear whether the memorized position that underlies the priming effect is coded relative to the observer (i.e. the egocentric position) or relative to other items in the search array (i.e. the allocentric position). Past studies did not address this question and typically if in priming trial the position remained the same relative to other items on the screen, it was also presented at the same position on the screen relative to the observer. Maljkovic and Nakayama (1996) attempted to address this issue by comparing an allocentric condition, where the target remained at the same location relative to the other items in the display but occupied a different egocentric location, with a combined allocentric–egocentric condition, where both the allocentric location and the egocentric location of the target remained the same. However, their results were inconclusive: although the priming effect appeared to be marginally bigger when egocentric information was added, it remained unclear whether this effect would be statistically significant.

The aim of the current study was to investigate, using a visual search task, priming of target location when both allocentric and egocentric information are available, and to directly compare this with allocentric and egocentric priming. If egocentric representations cannot be stored, as suggested by the perception–action model, egocentric priming will not be observed and there will be no or little difference between priming in the allocentric condition and priming in the combined allocentric–egocentric condition. However, if egocentric representations can be stored, egocentric priming will be observed and this will have implications for the perception–action model.

Section snippets

Participants

30 naïve participants from the University of Durham took part in this experiment and received course credit. Ethical approval was obtained from the Psychology Research Ethics Committee at Durham University and participants gave informed consent. Participants all had normal or corrected-to-normal visual acuity.

Stimuli

Each trial consisted of two sets of stimuli. First a letter was presented at the centre of the screen. Participants had to foveate this letter and report its identity. The purpose of this

Results

All analyses are concerned with participants’ reaction times to make a target present or target absent decision. Incorrect answers and outliers (responses more than two standard deviations above or below the mean) were removed. All data were tested for normality using the Shapiro–Wilk statistic; the data were normal unless otherwise stated.

For each participant the smallest font size they could read when fixating on it was established prior to the experimental trials (8 participants used font

Discussion

The aim of this experiment was to investigate location priming for targets defined in allocentric, egocentric, and combined allocentric–egocentric coordinates. Contrary to the predictions made by the perception–action model, priming was observed for all three conditions. Surprisingly, we observed that egocentric information not only makes a contribution to the combined priming effect but that it seems to be more effective than the allocentric information: priming in the combined

References (37)

  • C.G. Christou et al.

    View dependence in scene recognition after active learning

    Memory and Cognition

    (1999)
  • G. Committeri et al.

    Reference frames for spatial cognition: Different brain areas are involved in viewer-, object-, and landmark-centered judgements about object location

    Journal of Cognitive Neuroscience

    (2004)
  • V.A. Diwadkar et al.

    Viewpoint dependence in scene recognition

    Psychological Science

    (1997)
  • D. Elliott et al.

    The influence of premovement visual information on manual aiming.

    The Quarterly Journal of Experimental Psychology

    (1987)
  • C.A. Finlay et al.

    Updating representations of learned scenes

    Psychological Research

    (2007)
  • G. Galati et al.

    The neural basis of egocentric and allocentric coding of space in humans: A functional magnetic resonance study

    Experimental Brain Research

    (2000)
  • A. Hilstrom

    Repetition effects in visual search

    Perception & Psychophysics

    (2000)
  • L. Huang et al.

    Repetition priming in visual search: Episodic retrieval, not feature priming

    Memory & Cognition

    (2004)
  • Cited by (38)

    • Do we have distinct systems for immediate and delayed actions? A selective review on the role of visual memory in action

      2018, Cortex
      Citation Excerpt :

      Further supporting the notion that egocentric target information can be maintained for a considerable amount of time after the movement target is removed from view, Ball, Smith, Ellison, and Schenk (2009) reported that participants benefited from the availability of egocentric cues in a visual search task even after a delay of more than 2 s. They used the spatial priming paradigm of visual search where the repeated presentation of the search target at the same location will lead to reliable reduction of search time It was found that regardless of whether this position was coded in an allocentric or egocentric format the priming effect always survived delays of several seconds (Ball, Smith, Ellison, & Schenk, 2010; Ball et al., 2009). In conclusion, these findings are inconsistent with the idea that egocentric representations are highly transient.

    • Comparing the effect of temporal delay on the availability of egocentric and allocentric information in visual search

      2017, Behavioural Brain Research
      Citation Excerpt :

      By presenting a target in the same location across trials we inferred through a reduction in search times whether participants still had access to the target location information from the preceding trial [31,32]. Using a feature search task we observed that search times decreased when the location of the target relative to a ‘landmark’ in the display (allocentric priming) was repeated and when the location of the target relative to the observer (egocentric priming) was repeated across trials [33]. In two subsequent studies, both of which employed the same conjunction search paradigm used here, we investigated the nature of the egocentric frame of reference further.

    • Gaze-centered spatial updating in delayed reaching even in the presence of landmarks

      2013, Vision Research
      Citation Excerpt :

      In the present study, delays of up to 12 s had no influence on the gaze-dependent pattern of pointing errors when landmarks were present, which is in agreement with our previous study where we observed the same effect without landmarks but the same amounts of delay (Fiehler, Schütz, & Henriques, 2011). This result is also consistent with studies on visuo-spatial perceptual tasks showing that egocentric representations can persist over seconds (Ball et al., 2009, 2010). Thus, our findings do not support a sudden switch from an egocentric representation used for on-line movement control to an allocentric representation used for memory-guided movements when the goal of the action is no longer visible (cf., Westwood & Goodale, 2003).

    View all citing articles on Scopus
    View full text