The Time Course of Visual Processing: From Early Perception to Decision-Making.(Statistical Data Included)

Abstract

* Experiments investigating the mechanisms involved in visual processing often fail to separate low-level encoding mechanisms from higher-level behaviorally relevant ones. Using an alternating dual-task event-related potential (ERP) experimental paradigm (animals or vehicles categorization) where targets of one task are intermixed among distractors of the other, we show that visual categorization of a natural scene involves different mechanisms with different time courses: a perceptual, task-independent mechanism, followed by a task-related, category-independent process. Although average ERP responses reflect the visual category of the stimulus shortly after visual processing has begun (e.g. 75-80 msec), this difference is not correlated with the subject's hehavior until 150 msec poststimulus.

How do we perceive and understand a natural scene? What are the mechanisms involved in the extraction of its meaning, and how do they temporally relate to each other? Current theories of visual processing suggest a distinction between two different mechanisms: a perceptual process extracting information about different properties of the visual input, followed by a higher-level decision process evaluating the relevance of this visual information, in terms of the goals and expectations of the subject, in order to prepare and generate the appropriate behavioral response (Romo & Salinas, 1999; Schall & Thompson, 1999; Shadlen & Newsome, 1996). How these two mechanisms can be dissociated in time and space is a fundamental issue that is not easy to address (Nichols & Newsome, 1999; Schall & Thompson, 1999). In humans in particular, using reaction times as a dependent variable precludes the separation of the respective durations of perception, decision, and motor response stages (Luce, 1986).

Event-related potentials (ERPs), which can be recorded even in the absence of a behavioral response, have been shown to reflect high-level properties of the visual stimulus such as its identity or category after roughly 150 msec (Mouchetant-Rostaing, Giard, Bentin, Augera, & Pernier, 2000; Allison, Puce, Spencer, & McCarthy, 1999; Schendan, Ganis, & Kutas, 1998; Thorpe, Fize, & Marlot, 1996; Botzel, & Schulze, Stodieck, 1995). This holds for faces Jeffreys 1996) as well as other objects (Rossion et al., 2000). More recently, magnetoencephalography (MEG) has demonstrated a similar pattern of results (Halgren, Raij, Marinkovic, Jousmaki, & Hari, 2000). Likewise, experiments using electrophy0siological recordings in monkeys have reported neural responses highly selective to a specific visual category even before 100 msec (Vogels, 1999; Oram & Perrett, 1992; Perrett, Rolls, & Caan, 1982). A number of recent ERP studies in humans (Mouchetant-Rostaing et al., 2000; Debruille, Guillem, & Renault, 1998; George, Jeme l, Fiori, & Renault, 1997; Seeck et al,, 1997) seem to suggest, however, that high-level properties of the visual stimulus could be extracted much faster, sometimes reporting "face recognition" (Seeck et al., 1997) or "face gender discrimination" (Mouchetant-Rostaing et al., 2000) effects as early as 50 msec poststimulus.

However, the fact that neural activity varies with the properties of the visual input is not sufficient to conclude that the subject can actively recognize the identity of the categories involved. For example, it could simply reflect differences in the low-level properties of the visual stimuli, differences that are difficult if not impossible to investigate systematically.

Here we present an alternating dual-task paradigm ERP experiment that allowed us to compare the processing of the same visual category having different task-related status, and conversely, different visual categories having the same behavioral status. We were able to dissociate (i) a low-level sensory analysis activity and (ii) a high-level task-related mechanism that was clearly independent of the sensory properties of the stimulus.

RESULTS