Mark Beeman - US grants

Although the Left Hemisphere (LH) appears dominant for processing language, a great deal of evidence indicates that the right Hemisphere (RH) also processes language. Predictions from a new theory of RH language processing will be tested in a five-year program of experiments. This theory proposes that when processing a work the RH weakly activates a large "field" of related semantic representations, including peripherally related information. However, the activate is too diffuse to select information for further processing. This is termed coarse semantic coding. In contrast, the LH strongly activates a narrow semantic field of closely related information, enabling it to select representations for further processing or for consciousness. This is termed fine semantic coding. This theory predicts that RH language processing will be maximized in experiment tasks that a) utilize peripheral information; b) utilize overlap from multiple inputs; and/or c) not require selection of a single unique response. The proposed methods will equate performance of controls and patients with RH or LH lesions, and of normal subjects processing linguistic stimuli directed to each hemisphere, revealing qualitatively different patterns of performance. In the proposed project, 24 experiments will test the following general hypotheses: 1) That drawing coherence inferences depends on the RH homologue to Wernicke's area, because this area is important for recognizing distant relations between words. 2) That the semantic activation of peripherally related information depends on this same ares, and as a result the RH is more sensitive to particular types relations between words. 3) That the ease with which people can read words from each part of speech is modulated by differences in the semantic information conveyed by these words (as seen in alexic patients), and that the RH is more sensitive than the LH to these intrinsic word differences. 4) That subthreshold semantic activation in the RH is too weak to elicit explicit identification or awareness, but can influence binary decisions about broad semantic categories ("Is it a fruit") . 5) That the LH includes a layer of processing absent in the RH. The RH coarse semantic coding theory is pitted against two alternatives: That attentional mechanisms focus activation more sharply in the LH than in the RH; Or, that homotopic interhemispheric inhibition shapes the semantic activation of both hemispheres. According to the proposed theory, complete discourse processing requires coordination of both RH coarse semantic coding and LH fine semantic coding. The proposed project is expected to enhance knowledge of the specific brain regions and mechanisms underlying language processes. Such knowledge will inform future models and computer simulations of language processing, neuroscientific examination of anatomical asymmetries, and clinical recognition of the language impairments of patients with RH lesions and the residual language abilities of alexic patients with LH lesions.

? DESCRIPTION (provided by applicant): Countless anecdotes report sleep-assisted insight, or sudden solving of previously vexing problems. Prior research indicates that both sudden insight solutions (Wagner et al., 2004) and restructured memory (Ellenbogen et al., 2007) benefit from periods of sleep, possibly dependent on the same sleep- dependent processes (Stickgold & Walker, 2013). The proposed project adapts a recently developed memory paradigm that selectively enhances sleep-related memory consolidation to manipulate the incubation of problems during slow wave sleep. Thus, we can make more direct causal inferences about the sleep stages and processes that support incubation and subsequent solving of insight-like problems. If successful, the targeted incubation paradigm will provide a method for facilitating problem solving and for further investigating mechanisms and interactions of problem incubation and memory. The primary aim of the project is to test whether sleep facilitates problem solving through reactivation and restructuring of the problem representation during sleep. We test this by adapting the targeted memory reactivation paradigm developed by co-investigator Paller (e.g., Rudoy et al., 2009). Each participant engages in an evening and morning session for 5 days. Each evening session, participants attempt to solve a set of difficult insight-like problems, until 4 problems remain unsolved. The problems presented on any given day are distinctive by content, type, and modality (spatial versus verbal, etc.). During these evening sessions, each problem attempt is paired with a distinctive sound. Overnight, participants' sleep will be monitored with a portable EEG system capable of detecting sleep stages (particularly slow wave sleep). The sound cues for half the unsolved problems will be presented during slow wave sleep phases, triggering reactivation (Antony ... Paller, et al., 2012). The following morning, participants will re-attempt all the unsolved problems; we predict they solve more problems that were targeted with sound cues during sleep, because memory reactivation permits restructuring necessary to solve such problems. Follow-up experiments will examine other sleep stages, problem types, and the pre-sleep processing necessary for successful incubation. This study is novel and transformative, generating a novel paradigm for research and for facilitating problem solving in the real world. If successful, it will lead to a new program of research that promises t illuminate the important role of sleep in problem solving and in cognition generally, as well as mechanisms of interactions between memory and problem solving.