A functional MRI study of preparatory signals for spatial location and objects
Introduction
When looking at a visual scene, we can focus attention on the location or the identity/features of objects (e.g. whether it is a face or a building), and both forms of selection may occur during normal vision. A growing number of brain imaging studies have implicated a set of dorsal frontoparietal areas, including the putative human homologues of monkey areas LIP and FEF in the control of visual attention (reviewed in Corbetta & Shulman, 2002; Kanwisher & Wojciulik, 2000; Pessoa, Kastner, & Ungerleider, 2003). Activity in these areas is time-locked to the onset of cue stimuli that instruct subjects to direct and maintain attention to a location, or switch attention between locations, features, or whole objects (Corbetta, Kincade, Ollinger, McAvoy, & Shulman, 2000; Hopfinger, Buonocore, & Mangun, 2000; Kastner, Pinsk, De Weerd, Desimone, & Ungerleider, 1999; Liu, Slotnick, Serences, & Yantis, 2003; Shulman et al., 1999, Yantis et al., 2002). It has been proposed that these areas selectively bias visual cortex so that it responds more strongly to behaviorally relevant objects. For example, when expecting a certain object at a certain location, dorsal frontoparietal areas may bias activity, prior to stimulus presentation, in visual areas that code for the attended location and object.
Although studies indicate that preparatory signals in some dorsal areas do code selectively for attended visual attributes (e.g. location, direction of motion) (Bisley & Goldberg, 2003; Corbetta, Kincade, & Shulman, 2002; Hopfinger et al., 2000; Shulman, D’Avossa, Tansy, & Corbetta, 2002), the presence of selective activity is usually accompanied by a much more widespread activation of posterior parietal and frontal cortex, which appears to be general and not selective. The generality of frontoparietal recruitment during visual attention tasks has led to the proposal that activity in this network correlates with general factors such as perceptual difficulty or load (Marois, Chun, & Gore, 2000; Wojciulik & Kanwisher, 1999). Therefore, one goal of this experiment was to test the specificity of preparatory signals in dorsal fronto-parietal regions for location, by asking subjects to covertly attend to either a left or right visual field location.
Dorsal fronto-parietal areas are thought to control and modulate the inflow of visual sensory information from particular locations by sending biasing signals to regions in occipital cortex. In support of this idea, Kastner et al. reported a positive change of the blood oxygenation level dependent (BOLD) signal in visual cortex (both striate and extrastriate) prior to the presentation of a test array, consisting of complex visual pictures, at an expected location (Kastner et al., 1999). This expectation-related signal increase, known as a baseline signal, appeared to be location-selective since it was only seen in retinotopically appropriate regions of visual cortex (see also Hopfinger et al., 2000; Ress, Backus, & Heeger, 2000). However, in an experiment in which the location of a stimulus was cued, we failed to observe baseline shifts in visual cortex that varied with the direction of the cue, even though we observed behavioral effects of the cue as well as cue-related modulations in dorsal fronto-parietal regions (Corbetta et al., 2000). These results raise the question of whether behaviorally effective cues that modulate parietal cortex necessarily produce preparatory spatially specific biasing signals in retinotopic visual cortex.
Therefore, a second goal of the experiment was to revisit the issue of spatially-specific baseline signals in retinotopic visual cortex using a paradigm that was more similar to Kastner et al. (e.g. a match-to-sample task involving complex objects) while providing additional controls for the distribution of attention in the visual field. If baseline signals in retinotopic regions of visual cortex are location-specific, then they should move with the attended visual location. Since Kastner et al. cued the same location on each trial, they were not able to test this prediction.
While previous studies of attention have examined the specificity of dorsal fronto-parietal areas for simple features such as location or motion direction, less work has been directed at more complex classes of objects. Studies have indicated that regions in fusiform (‘fusiform face area’, FFA (Kanwisher, McDermott, & Chun, 1997)) and parahippocampal (‘parahippocampal place area’, PPA (Epstein & Kanwisher, 1998)) cortex show selective sensory-evoked responses to faces and places, respectively. Moreover, responses in FFA are modulated according to whether the task requires face or place discriminations (Wojciulik, Kanwisher, & Driver, 1998). If preparatory signals in the dorsal network also reflect object coding, then the spatial distribution or magnitude of preparatory activity may vary with the type of object. Conversely, if the dorsal network is primarily involved in selecting the location of objects, then similar preparatory signals should occur when either a face or an outdoor scene is expected. Therefore, a third goal of the experiment was to test for the specificity of biasing signals in dorsal fronto-parietal regions for different classes of objects, by asking subjects to perform object recognition tasks using unfamiliar faces or outdoor scenes.
Finally, task-specific modulations in FFA and PPA to faces and houses may result from preparatory biasing signals in fusiform cortex, analogous to the spatial biasing signals observed in retinotopic cortex. Therefore, a fourth goal was to examine whether selective, preparatory biasing signals could be observed in object-specific extrastriate regions. By cueing specific types of objects at particular spatial locations, we tested whether preparatory baseline shifts occur both in early retinotopic areas that process information presented at the cued location and in higher-order extrastriate regions that generalize over location but selectively process the object-type (e.g. face versus house) that has been cued.
Section snippets
Subjects
Twenty-one subjects were recruited from the Washington University community for experiment 1. Four subjects, all of whom had participated in experiment 1, participated in experiment 2. Informed consent was obtained in accordance with procedures approved by the local human studies committee. All subjects were strongly right-handed as measured by the Edinburgh Handedness Inventory, had normal or corrected-to-normal vision, and normal neurological history.
Apparatus and stimuli
Stimuli were projected using an Epson
Behavioral performance
Accuracy and RT data were analyzed with a 2 × 2 within-subject ANOVA with spatial cueing (face directional, face neutral), and discrimination (match, no-match) as factors. Responses were averaged across locations since no significant differences were found in accuracy or reaction times (RTs) between the two fields. Subjects used the spatial cue to solve the task. They were more accurate and faster in the face discrimination task when the arrow cue pointed to one location (directional cue) than
Visual field selectivity of preparatory signals for spatial attention in parietal regions
A rather puzzling observation, replicated now several times, is that preparatory signals in dorsal frontoparietal areas for spatial attention show weak evidence of location-selectivity during tasks that force subjects to either maintain covert attention on a location or switch attention between locations (e.g. Corbetta et al., 2000, Yantis et al., 2002). This result is puzzling since one would expect spatial attention signals to operate within a spatial map coding for extrapersonal locations,
Acknowledgements
MH71920-06, and the J.S. McDonnell Foundation.
References (53)
- et al.
Increased activity in human visual cortex during directed attention in the absence of visual stimulation
Neuron
(1999) - et al.
Hemispheric specialization in human dorsal frontal cortex and medial temporal lobe for verbal and nonverbal memory encoding
Neuron
(1998) - et al.
Neural correlates of the attentional blink
Neuron
(2000) - et al.
Separating processes within a trial in event-related functional MRI II. Analysis
NeuroImage
(2001) - et al.
A homogeneity correction for post-hoc ANOVAs in fMRI
NeuroImage
(2000) - et al.
Separating processes within a trial in event-related functional MRI I. The method
NeuroImage
(2001) - et al.
The generality of parietal involvement in visual attention
Neuron
(1999) - et al.
Functional organization of human intraparietal and frontal cortex for attending, looking, and pointing
Journal of Neuroscience
(2003) - et al.
Neuronal activity in the lateral intraparietal area and spatial attention
Science
(2003) - et al.
Primate frontal eye fields. I. Single neurons discharging before saccades
Journal of Neurophysiology
(1985)