Academic journal article Cognitive, Affective and Behavioral Neuroscience

Interactions between the Dorsal and the Ventral Pathways in Mental Rotation: An fMRI Study

Academic journal article Cognitive, Affective and Behavioral Neuroscience

Interactions between the Dorsal and the Ventral Pathways in Mental Rotation: An fMRI Study

Article excerpt

In this fMRI study, we examined the relationship between activations in the inferotemporal region (ventral pathway) and the parietal region (dorsal pathway), as well as in the prefrontal cortex (associated with working memory), in a modified mental rotation task. We manipulated figural complexity (simple vs. complex) to affect the figure recognition process (associated with the ventral pathway) and the amount of rotation (0° vs. 90°), typically associated with the dorsal pathway. The pattern of activation not only showed that both streams are affected by both manipulations, but also showed an overadditive interaction. The effect of figural complexity was greater for 90° rotation than for 0° in multiple regions, including the ventral, dorsal, and prefrontal regions. In addition, functional connectivity analyses on the correlations across the time courses of activation between regions of interest showed increased synchronization among multiple brain areas as task demand increased. The results indicate that both the dorsal and the ventral pathways show interactive effects of object and spatial processing, and they suggest that multiple regions interact to perform mental rotation.

The focus of the present study was to investigate a relationship between the dorsal and the ventral streams in high-level visual cognition. We used fMRI to examine how large-scale networks of cortical regions associated with spatial rotation and perceptual encoding are modulated by two variables: the degree of mental rotation and the complexity of the figure being rotated. The main components of the network to be examined include the left and right intraparietal regions (part of the dorsal stream or the so-called where system), the left and right inferior extrastriate (IES) and inferior temporal regions (part of the ventral or what stream; see, e.g., Mishkin, Ungerleider, & Macko, 1983; Ungerleider & Haxby, 1994), and the frontal system, which has been implicated in studies of visual and spatial working memories.

The theoretical perspective is based on the hypothesis that fMRI-measured activation is a correlate of resource consumption in a resource-based computational architecture and that more difficult tasks tend to require the consumption of more resources (Just, Carpenter, & Varma, 1999). In this perspective, fMRI activation provides a measure of cognitive workload. As support for these hypotheses, several studies have shown that fMRI-measured activation systematically varies with manipulations of task difficulty (cognitive workload) in domains as diverse as sentence comprehension (e.g., Just, Carpenter, Keller, Eddy, & Thulborn, 1996; Keller, Carpenter, & Just, 2001), n-back tasks (e.g., Braver, Cohen, Jonides, Smith, & Noll, 1997; Grasby et al., 1994) and mental rotation (e.g., Carpenter, Just, Keller, Eddy, & Thulborn, 1999; Cohen et al., 1996;Tagarisetal, 1997). The general argument is that the neural implementation of a process requires physiological resources, such as the neuronal, circulatory, and glial processes, as well as structural connectivity that ensures the coordinated communication of the networks that subserve the visuospatial experiences. The fMRI activation measure is interpreted as assessing one facet of the involvement of the large-scale neural networks (Mesulam, 1990, 1998) that correlate with cognitive computations.

The study focuses on three cortical areas associated with visuospatial processing. The first, the where system, is a processing stream that projects from the occipital cortex to the parietal region and that participates in the computation of extrapersonal and personal spatial localization (e.g., Mishkin et al., 1983; Ungerleider & Haxby, 1994). This system is thought to be involved in mental rotation, because the task requires the computation of the spatial coordinates of the object being mentally rotated and its comparison with the represented coordinates of the target object. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.