Object representations in ventral and dorsal visual streams: fMRI repetition effects depend on attention and part-whole configuration

Abstract

The effects of attention and object configuration on the neural responses to short-lag visual image repetition were investigated with fMRI. Attention to one of two object images in a prime display was cued spatially. The images were either intact or split vertically; a manipulation that negates the influence of view-based representations. A subsequent single intact probe image was named covertly. Behavioural priming observed as faster button presses was found for attended primes in both intact and split configurations, but only for uncued primes in the intact configuration. In a voxel-wise analysis, fMRI repetition suppression (RS) was observed in a left mid-fusiform region for attended primes, both intact and split, whilst a right intraparietal region showed repetition enhancement (RE) for intact primes, regardless of attention. In a factorial analysis across regions of interest (ROIs) defined from independent localiser contrasts, RS for attended objects in the ventral stream was significantly left-lateralised, whilst repetition effects in ventral and dorsal ROIs correlated with the amount of priming in specific conditions. These fMRI results extend hybrid theories of object recognition, implicating left ventral stream regions in analytic processing (requiring attention), consistent with prior hypotheses about hemispheric specialisation, and implicating dorsal stream regions in holistic processing (independent of attention).

Fig. 1

(A). Examples of intact and split images used in the current study and graphic representation of the experimental design. Both attended and uncued objects prime their corresponding probe when shown in an intact configuration (holistic priming, depicted by contrast-reversal), resulting in a main effect of configuration. Both intact as well as split objects prime an intact probe object when attended (analytic priming, depicted as cylinders), resulting in a main effect of attention.(B) Sequence of events within one trial of the main experiment.

Fig. 2

Behavioural priming, i.e., unprimed–primed RTs, for each condition of interest (AttInt = attended, intact condition, AttSpl = attended, split condition, UncIn = uncued, intact condition, UncSpl = uncued, split condition). Error bars reflect one-tailed 95% confidence intervals.

Fig. 3

Maximal intensity projections (MIP) of clusters that survived p < .05 corrected for their spatial extent (using a height threshold of p < .001 uncorrected) in the localiser session for the contrasts of: (A) objects (averaging over intact and split) vs scrambled objects, and (B) split vs intact objects.

Fig. 4

Voxel-wise results. On the left, statistical parametric maps of regions implicated in object recognition shown on orthogonal sections through a normalised structural of one participant. (A) Voxels in cyan were more active for objects than scrambled objects in the localiser (Fig. 3), and formed the search space for the voxels in red that showed a main effect of attention on repetition effects in the main experiment, thresholded at p < .001 uncorrected. Cross-hair centred on the left fusiform maximum that was the only maximum to survive correction for the search region. (B) Voxels in yellow were more active for split than intact objects in the localiser (Fig. 3), and formed the search space for the voxels in red that showed a main effect of configuration on repetition effects in the main experiment, thresholded at p < .001 uncorrected. Cross-hair centred on the right intraparietal maximum that was the only maximum to survive correction for the search region. (C) Voxels in cyan were more active for objects than scrambled objects in the localiser (Fig. 3), and formed the search space for the voxels in red that showed an interaction between attention and configuration on repetition effects in the main experiment, thresholded at p < .001 uncorrected. Cross-hair centred on the left lateral occipital maximum that was the only maximum to survive correction for the search region. On the right, percentage BOLD signal change between the peak of the fitted event-related response to the primed conditions subtracted from that of the corresponding unprimed conditions (i.e., where positive = repetition suppression (RS), and where percentage is relative to the mean BOLD signal over all voxels and volumes), for each of the four conditions of interest (AttIntact = attended, intact condition, AttSplit = attended, split condition, UncIntact = uncued, intact condition, UncSplit = uncued, split condition). Error bars are two-tailed 95% confidence intervals of repetition effects vs zero.

Fig. 5

Group-fROI results (A) On the left, BOLD repetition effects for attended and uncued conditions (averaged across intact and split configurations) for left (L) and right (R) fROIs within the ventral visual stream (averaged across posterior and anterior fROIs; see text). See Fig. 4 legend for further details. On the right, scatter plot of each participant's behavioural priming for attended primes (averaged across intact and split configurations) against RS in left fROI regions (averaged across posterior and anterior fROIs). (B) On the left, BOLD repetition effects for intact and split configuration conditions (averaged across attended and uncued) for left (L) and right (R) fROIs within the dorsal visual stream (see text). On the right, scatter plot of each participant's behavioural priming against RE for uncued, intact primes in intraparietal fROIs (averaged across left and right fROIs).