Verbs and nouns in awake neurosurgery: needs and answers

Awake neurosurgery (henceforth, awake surgery) is a medical procedure recommended for patients with brain tumors or pharmacologically intractable epilepsy. It borrows its name from the intraoperative stage in which patients are awake, so that neuropsychological and/or language tasks can be administered (e.g., Berger 1996; Duffau et al. 1999; Huncke, Van de Wiele, Itzhak, & Rubinstein, 1998; Kilbride, 2013; Ojemann & Mateer, 1979; Penfield & Boldrey, 1937). Awake surgery has gained popularity over the last years, because in more classical approaches - where patients are not awake and language tests cannot be administered - less tumor tissue may be removed and worse cognitive outcomes may be obtained (see, for a meta-analysis, De Witt Hamer, Gil-Robles, Zwinderman, Duffau, & Berger, 2012). Language testing is usually also performed shortly before and after surgery (the so-called “perioperative stages”). Awake surgery finds its way within a multidisciplinary field that includes a range of specialists – radiologist, anesthesiologist, neurosurgeon, neuropathologist, neurophysiologist and neuropsychologist (or speech/language therapist or clinical linguist). All these professionals have a specific role in the assessment and treatment of each individual patient. As researchers, we should strive to understand the needs of these professionals, and provide answers that help them reach their goals, while enabling us to gain knowledge on language and the brain, and to improve current standards of patient care. A recurrent topic throughout this dissertation is that the language capacities of patients with brain tumors have been traditionally assessed with few tasks, and sometimes with non-standardized materials (e.g., Finch & Copland, 2014; Rofes & Miceli, 2014). Object naming tasks have been a key component of these assessments (for a review see De Witte & Mariën, 2013). Object naming has much to commend it, as seeing a picture and producing a noun (“apple”) or a determiner phrase (“the apple”) engages lexico-semantic input and output processes that are indispensable for everyday communication (Goodglass & Wingfield, 1997). For example, perioperative processing speed on object naming has been shown to correlate with postsurgical chances of return to work in patients undergoing this type of surgery (Moritz-Gasser, Herbet, Maldonado, & Duffau, 2012). Furthermore, object naming tasks were introduced by pioneers in the field and have maintained their role as a gold standard in the identification of areas relevant for language processing during surgery (Hamberger & Tamny, 1999; Ojemann & Mateer, 1979; Ojemann, Ojemann, & Lettich, 2002; Sanai & Berger, 2009). Despite these positive results, having object naming as the main component of assessment in awake surgery may not be always sufficient. This is because (1) the language capacities required for object naming may be relatively spared in spite of damage to other language abilities such as reading, writing, comprehension, or naming of actions (Santini, Talacchi, Squintani, Casagrande, Capasso, & Miceli, 2012; Satoer et al., 2014); (2) there are language processes typically used in everyday language (e.g., for sentence formulation, to refer to actions, or to moments in time) that cannot be assessed with object naming (e.g., Rofes & Miceli, 2014; for a review); and (3) the neural correlates of object naming may be partially segregated from those of other processes such as action-naming (e.g., Mätzig, Druks, Masterson, & Vigliocco, 2009; Vigliocco, Vinson, Druks, Barber, & Cappa, 2011; for reviews). These are important issues in aphasiology, neurolinguistics, and cognitive neuroscience, which we revisited to work on critical improvements for awake surgery: the use of standardized tasks for language mapping, more specifically, the use of verbs as isolated words and finite verbs in sentences. Verbs differ from nouns in a number of ways: they typically refer to actions (Gentner, 1982), and may weigh differently on psycholinguistic variables that affect performance in individuals with aphasia and non-brain-damaged people (e.g., Whitworth, Webster, & Howard, 2005; Vigliocco, Vinson, Druks, Barber, & Cappa, 2011). Verbs also entail a predicate-argument and a thematic structure (Carlson & Tanenhaus, 1989), and when they are used in the context of a sentence they require that agreement relations between the subject and the verb be fulfilled (Hale & Keyser, 1998). Verbs also allow the positioning of events in time and reference to events within a specific time frame (Bastiaanse, Bamyaci, Hsu, Lee, Yarbay Duman, & Thompson, 2011). Furthermore, they may be processed in different brain regions compared to nouns. It has been shown that verb processing elicits activation in the frontal and inferior-parietal lobe, whereas nouns trigger more activation in the temporal lobe (e.g., subjects KJ-1360, AN- 1033 and Boswell in Damasio & Tranel, 1993; Lubrano, Filleron, Démonet, & Roux, 2014; Miceli, Silveri, Noncentini, & Caramazza, 1988). Overall, earlier research suggests that the current assessments in awake surgery are pertinent but not sufficient. In this dissertation, we contribute to the efforts that are underway to meet the needs of awake surgery, both in terms of behavioral and anatomical knowledge. We approach the issue from the perspective of aphasiology, neurolinguistics and cognitive neuroscience (e.g., Coltheart 2011; Withworth et al., 2005). The chapters of this thesis each correspond to an article published in an international peer-reviewed journal or to a final draft that is (or will be) submitted for publication. The thesis begins with a review of language mapping with tests that use verbs and sentences (Chapter 2). This chapter serves as a critical introduction to some of the needs that neurosurgeons and people responsible for the assessment of language skills in individuals with brain gliomas face in everyday practice. More specifically, it talks about renewing the consideration of tasks that use verbs, an approach already tackled in classical papers such as Ojemann and Mateer (1979). After this introduction, we deal with the design and development of an object naming task and a task with finite verbs in sentences, specifically designed to assess Italian-speakers undergoing awake surgery (Chapter 3). Such tools did not exist for the Italian language. The standardization of the tasks is described so that both the task methodology and the standardization of the tasks can be easily adopted by other research or awake surgery teams. In Chapter 4, the validity of the noun and verb picture-naming tasks are assessed, as well as two other tasks typically used to assess language deficits in communicative contexts. By evaluating the performance of people suffering from post-stroke aphasia, we assess how well these naming tasks correlate with the ability to use language in an everyday context. The next two chapters describe the implementation of these tasks in surgical rooms and report the neurofunctional results of perioperative tests, their relation to the current neuroscience of language literature, and a critical review of the advantages and disadvantages that language mapping tasks may entail for awake surgery (Chapter 5 and Chapter 6). We finish the dissertation with a general discussion, aimed at describing theoretical and clinical reasons way a rigorous approach may improve awake surgery procedures. We also discuss some future implications of this line of work (Chapter 7).