| Building the bridge between perception and memory |
 |
 |
 |
|
Even when the external object of perception has departed, the impressions
Lorem ipsum dolor sit amet, consectetur adipiscing elit. (Bauman et al, 2008)
The study of visual perceptual memory, a field of perception and memory research initiated 20 years ago (see reviews by Magnussen, 2000; Magnussen, Greenlee, Baumann & Endestad, 2009), is an ideal starting point for such a bridge building because of its experimental simplicity and its firm anchoring in theories of visual information processing. The break-through that forms the basis of the present project, is the result of a recent fMRI study (Baumann, Endestad, Magnussen & Greenlee, 2008), in which we provided strong evidence that retrieval of information from perceptual memory involves the regeneration of activity in early visual areas of the brain, supporting specific models of visual perceptual memory (Magnussen, 2000). This led to the formulation of a more general project in visual memory, governed by a conceptual framework resting on two theoretical pillars: First, in general agreement with a multiple-systems concept of memory and the view that memory performance may be supported by several memory systems acting in concert, we assume the existence of parallel visual memory systems. The second assumption is that the memory of an object, scene or event, involves the recruitment of the very perceptual processes that took place at the time the object, scene or event, was experienced. On this account, a visual representation is indeed a “re-presentation” of the perceptual aspects of the object or scene. Part 1 of the project examines the time of transfer of information from perception to memory. It is assumed that information is maintained in a perceptual representation mode for some time -- of the order of seconds rather than milliseconds -- before it is transferred to a memory mode, and that by recording of response times (RT), it is possible to delineate perception and memory processes (Magnussen et al., 2009). Part 2 of the project investigates the read-out from visual memory. An unresolved question is whether read-out from visual memory involves reactivation of perceptual processes and the recruitment of brain regions involved in the on-line analysis of visual information. And if so, how early in the processing hierarchy does recruitment take place? Here, the results of the cognitive experiments will guide the design of fMRI experiments to decide on the transition between perception and memory processes, and the overlapping and non-overlapping brain regions involved in perceptual and memory processing. The delayed discrimination task solves a major problem in designing fMRI experiments (Baumann et al., 2008): in delayed discrimination the only difference between the two tasks is the time interval between the stimuli to be compared – in all other respects they are identical. The results from a recent fMRI study of memory masking (Sneve, Endestad, Alnæs, Greenlee & Magnussen, 2011) are encouraging. Part 3 of the proposed project applies a novel methodology to the study of visual memory for pictorial information. While the human visual system may be able to extract very rapidly (“in a single glance”) meaningful information about the whole scene (i.e., its “gist”), information about pictures is acquired under normal viewing conditions by sampling pictorial content or a real-scene content in successive fixations. There is evidence that the scanning pattern is related to the memory for pictures. In an innovative study from our laboratory, Laeng recorded eye movements while participants tried to memorize “The music lesson” by Vermeer, and when the participants viewed a blank screen and tried to recollect the painting (Laeng & Teodorescu, 2002). The results showed a remarkable similarity in eye movements and fixation sequences between the original scrutiny of the stimulus and those during recollection or image generation. This strongly suggests that eye scanpaths during visual imagery not only reenact those of perception of the same visual scene but they also play a functional role. A novel method, called the “shifting probe” method, has been developed in our laboratory and consists of a circular probe that, at recognition, is presented in the same locations of each of the original fixations and shifts and dwells over each of these corresponding regions of the scene. Project team: Tor Endestad, Bruno Laeng, Svein Magnussen, Markus Handal Sneve og Dag Alnæs, CSHC. |
The overall aim of the present project is to close the gap in our understanding of visual perception and visual memory, focusing on the question of overlap and uniqueness of perception and memory processes. When -- and possibly where -- are perceptual representations transformed to memory representations, and to what degree does retrieval of information from visual memory involve regeneration of on-line visual processes? To answert these questions, the techniques of experimental cognitive psychology are combined with functional brain imaging (fMRI). Projects
Building the bridge between perception and memory