Forward Modeling in the Manual Modality: Linguistic and Nonlinguistic Predictions by American Sign Language Users
- Author(s): Brozdowski, Christopher Ryan;
- Advisor(s): Emmorey, Karen;
- et al.
Motor simulation has emerged as a mechanism for both predictive action perception and language comprehension. By deriving a motor command for an observed stimulus, and engaging in covert imitation, individuals can predictively represent the outcome of unfolding action as a Forward Model. In the context of these proposals, language is described as a highly systematized form of action that relies on the same simulation mechanisms. Some evidence also points to motor stimulation as a supplementary mechanism only under noisy or high-demand circumstances. Evidence of simulation can be derived from error patterns that defer to attributes of the predicting individual, as Egocentric Bias, or through differential responses to Symmetricity, one- vs. two-handed stimuli. Additionally, the sign language literature provides evidence that signers generate predictive representations based on information prior to the onset of a target sign. It is currently unclear, however, what features of transitions during the fluid sign stream make such predictions possible.
Experiment 1 examines the role of (a) motor simulation during action prediction, (b) linguistic status (i.e., pseudosigns vs. grooming gesture) in predictive representations and (c) language experience (i.e., signers vs. nonsigners) in generating predictions. As Egocentric Bias was only observed for non-linguistic stimuli, and only for nonsigners, Experiments 1 does not support strong motor simulation proposals and instead highlights the role of stimulus familiarity. The Experiment 2 focuses on movement and handshape as possible informative transitional information. While movement facilitated predictions regardless of language background and linguistic status of the stimulus, only sign language users relied on transitional handshape information and only for linguistic stimuli. Experiments 3 and 4 examine predictive processing through the lens of motor memory and motor imagery to further investigate the hypotheses that sign language users (a) only exhibit improved performance in linguistic contexts, and (b) are not sufficiently taxed by the present tasks to engage motor simulation during predictive processing. Only participants without sign language experience (a) showed evidence of using motor simulation, and (b) recruited memory and imagery abilities in generating predictive representations. While predicting the future can be difficult, sign language experience seems to shape how some predictions are made.