Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

III: Theory & Method in Functional Brain Imaging Progress in neurolinguistic research depends on the theoretical and methodological integrity of individual studies. To the extent that the research questions asked in each study are grounded in sound psycholinguistic theory and experimental methods, the results will make a valuable contribution to our understanding of how the brain processes language. In Unit 5, we reviewed literature that provides important findings about brain support for phonological, lexical-semantic, and syntactic operations. Until recently, it has been more common to find studies that, although ostensibly investigating specific aspects of language behavior (e.g., Glaser et al., 2013), lack in theoretical justification and/or methodological rigor. The following are just a handful of examples: Numerous studies have examined brain activations while participants listen to or read nouns vs. verbs. While this distinction relates to grammatical categories, and therefore, has some theoretical basis, the amount of research devoted to this distinction is disproportionate to its theoretical importance; also, the types of baseline conditions and the use of the subtraction method in many of these studies bring the findings into question. Ocklenburg (2013) investigated the relation of microstructural asymmetries in language-relevant white matter pathways and functional activation asymmetries during silent verb generation [when prompted by a letter, e.g., "B"] and when passively listening to spoken words [which started with the same letter]. Did this study really provide a clear window into speech production and speech perception, as claimed by the authors? Wang, et al. (2012), andAllendorfer et al. (2012), are recent studies (of a long history of studies) that have asked participants to generate verbs from nouns (e.g., presented with "hammer" participant is supposed to generate a verb such as "pound"). The use of such a task seems to be perpetuated by tradition, yet it lacks any sound theoretical rationale. Binder et al. (1997) wanted to examine neural support for semantics, so they had participants decide whether an animal is both native to the U.S.A. and used by humans. Embick et al. (2000) were interested in syntactic processing by the brain. They compared brain activation for sentences containing grammatical vs. spelling errors. Papathanassiou et al. (2000) assumed that they could “derive a functional definition of epicentre/essential language areas as regions implicated in both language production and comprehension tasks.” How did they go about investigating this “epicenter”? They performed a conjunction analysis of story listening and verb generation. Heim et al. (2003) set out to examine syntax and the brain. To do so, they asked participants to make decisions about: whether words are prepositions or nouns, the gender of nouns, and the spacing of consonant letter strings (baseline). Bartha et al. (2003) wanted to get at semantic processing in the brain. Participants were asked to decide whether concrete nouns are available in the supermarket and cost less than a certain amount of money! Thierry et al. (2008) tested the effect of word class conversion as used by Shakespeare – the functional shift – on event-related brain potential waves. Tesan et al. (2012) asked how the brain responds to any! In addition to the fact that tax dollars are allocated each year to support this kind of research, the publication of such studies militates against scientific progress in neurolinguistics (see Poeppel & Embick, 2005). Research with questionable theoretical and/or methodological foundations distracts and detracts from where our focus and energy should be devoted—namely, putting to test psycholinguistic empirical findings and the theories they support as these bear on language-brain relationships. What is the alternative, you may ask? Some examples of more worthwhile directions in different linguistic domains follow: Phonology: underspecification/markedness, segmental vs. metrical constraints Lexical: word neighborhoods, frequency Semantic: functional (e.g., tool, consumable) vs. sensory/motor features Syntactic: heuristic strategies (e.g., word order) vs. phrase structure Discourse: surface form (e.g., literal meaning) vs. inferences (e.g., figurative meaning) The following is an example of a study with a well thought-out methodological design and theoretical approach: Moseley, R. L., & Pulvermüller, F. (2014). An important matter to consider when designing any neurolinguistic study is the kind of experimental task one chooses to use. For more than two decades, psycholinguistic researchers have drawn a distinction between on- vs. off-line tasks. To date, most neurolinguistic studies have employed off-line tasks. In off-line tasks, responses are made after (rather than as) processing takes place and they engage conscious processing, leading to “extra-linguistic” demands on memory and other cognitive systems. An example of an off-line task is making a plausibility judgment on a sentence, or matching a heard sentence to one of several pictures, only one of which depicts the sentence meaning. Because data from off-line tasks reflect the combined demands of processing linguistic information and doing something further with that information, it is not possible to link the source of a person’s performance to one or the other of these stages of processing. On-line tasks, on the other hand, require a response that is made during initial processing and entail no conscious reflection on the stimulus. Consequently, performance can be linked directly to the demands associated with processing the stimulus. An example of an on-line task is the cross-modal priming paradigm employed by Burkhardt et al (2003). Understanding the distinction between on- and off-line measures of linguistic performance is critical when interpreting the underlying sources of an aphasic person’s impairments— that is, are they due to compromised linguistic processing or to post-linguistic processing limitations? The evidence thus far indicates that agrammatic aphasic persons experience a delay in activating information on-line, and this may lead to difficulty in producing or comprehending language. Conversely, on some tasks these individuals have shown more intact off- than on-line performance, suggesting that when given sufficient time they can “make up for” their slower automatic processing. Fluent aphasic persons seem to process information on-line in a timely fashion, but do not inhibit competing information. This results in poor on- and off-line task performance. All in all, it is clear that drawing conclusions and/or generalizations about localization of language solely from off-line tasks is not appropriate since such data reflect not only the operation of language components but also the other demands associated with the task. On the other hand, one advantage of employing off-line tasks is that they may tell us something about how individuals use language in communication contexts where reanalysis and reflection on what is said is the normal course. (for further discussion of the influence of task characteristics and demands, see Caplan & Gow, 2012). Conclusions: A critical examination of a neurolinguistic study will address both its theoretical rationale and its methodological design. More consistent attention to these characteristics is needed for the field to make significant progress in understanding how the brain and mind process language. IV: Review Cognitive processing constraints contribute to the different patterns of deficits/brain activation in language processing; Automatic, procedural, and timing/activation mechanisms and processes receive support from the inferior frontal cortex and subcortical networks; Controlled, declarative, and inhibition mechanisms and processes receive support from the posterior superior temporal and inferior parietal cortex; Employing theoretically motivated, innovative, and carefully controlled psycholinguistic experimental designs will advance our understanding of language and the brain.