Brain Networks Subserving the Extraction of Sentence Information and Its Encoding to Memory

Abstract

Sentences are the primary means by which people communicate information. The information conveyed by a sentence depends on how that sentence relates to what is already known. We conducted an fMRI study to determine how the brain establishes and retains this information. We embedded sentences in contexts that rendered them more or less informative and assessed which functional networks were associated with comprehension of these sentences and with memory for their content. We identified two such networks: A frontotemporal network, previously associated with working memory and language processing, showed greater activity when sentences were informative. Independently, greater activity in this network predicted subsequent memory for sentence content. In a separate network, previously associated with resting-state processes and generation of internal thoughts, greater neural activity predicted subsequent memory for informative sentences but also predicted subsequent forgetting for less-informative sentences. These results indicate that in the brain, establishing the information conveyed by a sentence, that is, its contextually based meaning, involves two dissociable networks, both of which are related to processing of sentence meaning and its encoding to memory.

Neural substrates of narrative comprehension and memory

“…Behaviorally, the use of situation models aids comprehension and memory of narrative text considerably Gernsbacher et al, 1990); however, the neural mechanisms that support this mnemonic boost have not been fully specified. While many fMRI studies have identified brain activation that predicts memory for individual words (Kirchhoff et al, 2000;Wagner et al, 1998) or sentences (Casasanto et al, 2002), only one previous study has identified memory-predictive activation in the context of coherent narratives (Hasson et al, 2007). The latter study found differential subsequent memory effects as a function of expectancy violation, but did not directly compare subsequent memory effects for coherent and incoherent materials.…”

“…Several fMRI studies have suggested that the right hemisphere is selectively involved in high-level discourse comprehension (Robertson et al, 2000;St George et al, 1999). Again, however, this conclusion is contradicted by other studies that have observed bilateral activation during story reading (Hasson et al, 2007;Maguire et al, 1999), or have found evidence for left hemisphere engagement in discourse processing using lateralized visual-field procedures (Prat et al, 2007). Thus, the precise relationship between narrative-level and sentence-level comprehension remains unclear.…”

“…Many of these regions are known to play a relatively general role in language processing -e.g., areas along the middle and superior temporal gyri and inferior frontal cortex (Binder et al, 1994;Ferstl and von Cramon, 2001;Huettner et al, 1989;Maguire et al, 1999;Robertson et al, 2000;St George et al, 1999), which show consistent recruitment in a broad range of word-level language tasks (Fiez and Petersen, 1998;Turkeltaub et al, 2002;Vigneau et al, 2006). However, other regions appear to be specifically recruited during comprehension of coherent text, including the anterior temporal lobes (ATL; Ferstl et al, 2007;Mazoyer et al, 1993;Stowe et al, 1998;Stowe et al, 2005) and dorsomedial prefrontal cortex (DMPFC; Hasson et al, 2007;Xu et al, 2005).…”

“…Most commonly, narrative-level mechanisms are isolated by contrasting brain activation when reading connected sentences or stories with activation when reading sentences or stories that are unrelated or inconsistent to varying degrees (e.g., Ferstl et al, 2005;Ferstl andvon Cramon, 2001, 2002;Fletcher et al, 1995;Giraud, 2000;Hasson et al, 2007;Vogeley et al, 2001;Xu et al, 2005). The results of such investigations converge on a distributed network of cortical regions subserving discourse-level comprehension.…”

“…Moreover, memory for narratives was assessed using only a verbatim recognition test; deeper comprehension of materials was not evaluated. Thus, it is unclear to what extent the subsequent memory effects observed by Hasson et al (2007) depend on narrative-level use of situation models vs. word-or sentence-level processes.…”