Adaptation of Inputs in the Somatosensory System

This chapter summarizes evidence that cortical maps and cortical response properties are in a permanent state of use-dependent fluctuations, where ‘‘use’’ includes trainingand learning-induced changes. In their simplest form, usedependent changes are input driven. Although attention and other high-level processes may contribute and enhance use-dependent neural changes by specific pathways conveying top-down information, reorganization can occur in the absence of high-level processes. The current experimental data imply that altered performance is based on altered forms of neural representations, and that all forms of perceptual learning can therefore be assumed to operate within the framework of cortical adaptivity. 2.1 Introductory Remarks 2.1.1 Two Forms of Plasticity Postontogenetic plasticity describes the capacity of adult brains to adapt to internal or environmental changes. It is useful to distinguish between two different forms of adult plasticity: 1. Lesion-induced plasticity, which subsumes cortical reorganization after injury or lesion, induced either centrally or at the periphery, refers to compensation for and repair of functions acquired before the injury or lesion. 2. Trainingand learning-induced plasticity, often called ‘‘use-dependent plasticity,’’ refers to plastic changes that parallel the acquisition of perceptual and motor skills. Because, for example, amputation changes the pattern of use entirely, a more accurate distinction would be between ‘‘lesion-induced’’ and ‘‘nonlesion-induced’’ plasticity. To what extent the two forms are based on di¤erent or perhaps even on similar mechanisms is a matter of ongoing debate. In contrast to developmental plasticity, adaptations of adult brains do not rely on maturation or growth. For learning-induced alterations, there is agreement on the crucial role played by so-called functional plasticity based on rapid and reversible modifications of synaptic e‰cacy, although largescale amputations have been shown to involve sprouting and outgrowth of a¤erent connections into neighboring regions at cortical and subcortical levels (Florence, Taub, and Kaas 1998; Jain et al. 2000). 2.1.2 Sites of Changes Perceptual learning is often highly specific to stimulus parameters such as the location or orientation of a stimulus, with little generalization of what is learned to other locations or to other stimulus configurations (see chapters 9, 11, 12, 14). Selectivity and locality of this type implies that the underlying neural changes are most probably occurring within early cortical representations that contain well-ordered topographic maps to allow for this selectivity (see chapter 1). In addition, a transfer of the newly acquired abilities is often considered an important marker of the processing level at which changes are most likely occur: limited generalization is taken as evidence for high locality of e¤ects in early representations. In contrast, transfer of learned abilities is taken as evidence for the involvement of higher processing levels often observed in task and strategy learning (see chapters 13, 14). There is increasing evidence that changes in early cortical areas might be more directly linked to perceptual learning than previously thought (Karni and Sagi 1991; Recanzone, Jenkins, et al. 1992; Schoups, Vogels, and Orban 1995; Crist et al. 1997; Fahle 1997, chapter 10). In fact, most of what we know today about adaptation of the somatosensory system comes from the investigation of the somatosensory areas characterized by extended and ordered neural representations of the body surface (box 2.1). In contrast, less is known about both the role of higher areas and the interaction between sensory association areas for perceptual learning. In any case, the conjecture that perceptual learning a¤ects early areas provides an important conceptual link to somatosensory adaptational processes (see chapters 9–14). 2.1.3 Driving Forces That Lead to Adaptational Changes What factors might induce changes in neural representations? Let us assume a dynamically maintained steady state of representations emerging from learning during development and adulthood that reflects the adaptation history to a ‘‘mean environment,’’ defined as the accumulated and idiosyncratic experience of an individual. Adaptational processes are assumed to operate on these representations, and long-lasting changes are likely to occur when sensory input patterns are altered such that they deviate from the mean environment. The average steady state can be altered in three principal ways: 1. By changing the input statistics. Specifically effective in driving adaptational changes are simultaneity, repetition, and, more generally, spatiotemporal proximity (see chapters 14, 20). Because these changes in input do not involve attention or processing for meaning, they induce a class of noncognitive adaptations based largely on bottom-up processing. 2. By drawing attention to certain aspects of a stimulus, thereby selecting it in comparison to others. The relevance of a stimulus can also change, depending on context, history, and behavioral task, thereby modifying how physically defined attributes are processed. There is general agreement that modification of early sensory processing by attention and stimulus relevance reflects top-down influences arising from cognitive processes (see chapters 13, 14, 20). 3. By using reward or punishment to reinforce learning. Such influences usually accelerate adaptational processes and are assumed to be mediated by specific brain regions modifying early sensory processing (see chapter 20). 2.1.4 The Hebbian Metaphor A central paradigm in the description and analysis of cortical plasticity is built around the Hebbian concept (1949): episodes of high temporal correlation between preand postsynaptic activity are prerequisite for inducing changes in synaptic e‰cacy. Historically, the idea that cooperative processes are crucially involved in generating long-lasting changes in excitability can be traced back to the nineteenth century ( James 1890). Indeed, since Hebb, the aspect of simultaneity has become a metaphor in neural plasticity, although the exact role of Hebbian mechanisms in use-dependent plasticity remains controversial (Carew et al. 1984; Fox and Daw 1993; Granger et al. 1994; Montague and Sejnowski 1994; Joublin et al. 1996; Buonomano and Merzenich 1996; Edeline 1996; Cruikshank and Weinberger 1996a,b; Ahissar et al. 1998). It has been suggested that the definition of Hebbian mechanisms 20 Hubert R. Dinse and Michael M. Merzenich