Counterfactual discourse in the brain. An ERP study with high-density montage. Mabel Urrutia & Manuel de Vega Universidad de La Laguna INTRODUCTION Frequently we listen or read counterfactual sentences such as: “If Mary had won the lottery, she would have bought a Mercedes car”. Understanding such kind of sentences seem to involve processing a double meaning, keeping in mind two alternative views of the situation. Thus, in the example we must be able to create a scenario in which Mary actually won the lottery and bought a Mercedes car. But at the same time, we must be aware that in fact Mary did not win the lottery or buy a car. In a previous study we explored the dual meaning of counterfactuals (de Vega et al. , in press). Readers were given short stories including an initial scenario, followed by a critical sentence that was either factual or counterfactual. A final sentence followed that was an appropriate continuation to one of the critical sentences. The data showed a clear consistency effect in the factual, but not in the counterfactual version of the stories. Namely, when the final sentence was a reinstatement of the initial situation (previous to the critical sentence), reading was considerably faster in the counterfactual than in the factual version, because updating occurs in the latter but not in the former stories. However, when the final sentence was a completion of the on-going situation, reading was equally fast in the factual and counterfactual stories, suggesting that a factual interpretation of counterfactuals is momentarily available, and confirming the idea of their dual meaning (e. g. , Roese, 2005; Byrne, 2002). The present research used an event related brain potentials technique (ERP), which could provide an optimal temporal resolution to explore the consistency effects both in factual and counterfactual stories. Thus, we will analyse the electrophysiological signature of the paradoxical dual meaning of counterfactuals. We employed a high-density montage, which allowed estimating tomographic inverse solutions, by means of the Variable Resolution Electromagnetic Tomography (VARETA) method (Bosh-Bayard, et al. , 2001). METHOD DESIGN & PROCEDURE A 2 Context (factual / counterfactual) 2 Final consistence (with factual / with counterfactual) factorial design. Forty participants were given 160 stories, 40 for each of the possible experimental versions. The stories were given auditorily, except the 6 -words final sentence, which was presented visually word-by-word, and participants had to answer comprehension questions in 30% of the stories. The 4 different versions for each story were counterbalance across-participants. EXAMPLE OF MATERIALS INITIALSENTENCE Marta wanted to plant flowers in her garden, but the soil was hard and dry. CRITICAL SENTENCE Because Marta found a spade she started to dig a hole [factual] If Marta had found a spade she would have started to dig a hole [counterfactual] FINAL SENTENCE She planted into the hole some roses [consistent with factual] She bought a spade in the market [consistent with counterfactual] FINAL POSITION WORD (e. g. , “roses”) (1 B) Counterfactual (1 C) Factual The ERP register was time-locked to the last word of the critical sentence (e. g. , hole) and the last word of the final sentence (e. g. , roses). The EEG was recorded from 64 thin electrodes mounted on an electrode cap and amplified using a Medicid-4 system. RESULTS FIRST POSITION WORD (e. g. , “hole”) (1 A) Counterfactual vs factual ERP RECORDING (2 A) Counterfactual Consistent vs Factual Consistent (2 B) Counterfactual consistent (2 C) Factual Consistent (3 A) Counterfactual Consistent (3 B) Counterfactual Consistent (3 C) Factual Inconsistent vs Factual Inconsistent First position word. There was a larger negativity in the 400 -500 window for the word following a counterfactual, although the effect was not statistically significant. However, the tomographic solution of VARETA showed a different distribution of brain activity. Particularly, an activation of the posterior cingulate region for counterfactuals, and of the orbito-frontal gyrus for factuals. Final position word. In the 350 -600 window there was a significant effect of Context (F(1, 34)= 4. 23, MSe= 192; p<. 05), namely the N 400 component was more negative in sentences following a counterfactual than a factual context (see Figure 2 A), as a N 400 consequence of the additional cognitive effort of processing counterfactual meaning. However, this effect was qualified by an interaction with Consistence (F (1, 34) = 6. 60; MSe=356, p<. 02). When factual stories were analyzed separately there was a main effect of Consistence (F (1, 34)= 7. 33; MSe= 173, p<. 01), with a larger N 400 in the inconsistent versions. This suggests an updating process, which makes it more appropriate to refer to the most recent factual events (consistent) than to the initial situation (inconsistent). By contrast, a separate analysis performed with counterfactual stories did not find Consistence effects (F (1, 34)= 2. 45, p=. 10). In other words, after reading a counterfactual the ERP pattern was similar for consistent and inconsistent sentences, suggesting that both the factual and the counterfactual meaning coexist. These results are compatible with the reading times data reported in the aforementioned study (de Vega, et al. , in press) and, therefore, with the dual meaning hypothesis. The VARETA solutions were quite distinctive for counterfactuals and factuals as we will comment later. CONCLUSIONS The experiment showed that sentences and words following a counterfactual sentence produce a different ERP signature than the same sentences and words following a factual sentence. In addition, tomographic estimations also differ considerably. Counterfactuals determine more complex process in the next subordinate sentence, as shown by the activity of the posterior cingulate region, which is usually associate to situation model updating, or the processing of temporal sequences of events. Counterfactuals determine much more processing load in the continuation (final) sentence as shown by a larger N 400 component, even when the sentence was consistent with the counterfactual meaning. In addition, the VARETA solution showed two important sources of brain activation, exclusively for counterfactuals. First, counterfactuals activate the left hemisphere: the superior temporal gyrus, the precentral gyrus, and the insula. This circuitry, specially the superior temporal gyrus and the insula has been related to emotional responses to moral judgment, or empathy (e. g. , Blair, 2005; Moll et al. , 2005). This confirms the idea that counterfactuals trigger complex emotions such as regret, frustration, and the like (e. g. , Byrne, 2005; Roese, 2002). Second, counterfactuals activate more the anterior and the posterior cingulate regions, which probably play a role in processing theory of mind and event sequences, respectively (e. g. , Ferstl & von Cramon, 2002; Saxe & Powell, 2006). The EEG based tomography is a very low resolution approach. Further research using high-resolution neuroimaging techniques (e. g. , f. MRI) will be necessary to explore the brain basis of the comprehension of counterfactuals with much more anatomical detail. Brain & Discourse Workshop. March, 2007. Leiden, Netherlands. REFERENCES (1) Blair, R. J. R. (2005). Responding to the emotion of others: Disociating forms of empathy through the study of typical and psychiatric populations. Conciousness and Cognition, 14, 698718. (1) (2) Bosch-Bayard, J. , Valdés-Sosa, P. , et al (2001) 3 D Statistical Parametric Mapping of EEG source spectra by means of Variable Resolution Electromagnetic Tomography (VARETA). Clinical Electroencephalography, 32, 47 - 61. (3) Byrne, R. M. J. (2005). The rational imagination. How people create alternatives to reality. Cambridge, M. A. : MIT Press. (4) De Vega, M. , Urrutia, M. , & Riffo, B. (in press). Cancelling updating in the comprehension of counterfactuals embedded in narratives. Memory and Cognition. (5) Ferstl, E. C. & von Cramon, Y. (2002). What does frontomedial cortex contribute to language Processing: Coherence or theory of mind? . Neuroimage, 17, 1599 -1612. (6) Moll, J. , Oliviera-Souza, R. , Tovar Moll, F. , Azevedo Ignácio, F. , Bramati, I. , Caparelli-Dáquer, E. , Eslinger, P. The moral affiliations of disgust. A functional MRI study. (2005). Cog Behav Neurol, 18, 68 - 78. (7) Roese, N. (2005). If only: How to turn regret into opportunity. New York: Broadway Books. (8) Saxe, R. , & Powell, L. J. (2006). It’s the thought that counts. Specific brain regions for one component of theory of mind. Psychological Science, 17, 692 -699
Скачать презентацию Counterfactual discourse in the brain An ERP study
d5d819a6366cd35158d15587dbb27dc1.ppt