Monday, 20 January 2014

Validation and Comprehension in Reading: Psychologists weigh in

The first 2014 edition of Discourse Processes is a special issue on the psychological process of validation and how it differs from comprehension. The 200-page volume presents 10 papers, 8 of which are original empirical studies, and the other 2 of which are bookends orienting us theoretically and historically on the theme. Definitions of the target concept vary but all involve a pairing of a new representation and world knowledge: “computing truth values or plausibility based on relevant world knowledge” (Isberner and Richter, 8), while the concluding chapter says validation is “the monitoring of information consistency during comprehension” (Kendeou, 189), and “it is not the activated information that is validated; it is the linkages between concepts/ideas that are validated” (Cook and O’Brien, 44). 

The core theoretical debate addressed is whether validation occurs after comprehension or simultaneously, i.e., on a two-step or one-step model. Is validation passive, non-optional, and performed through asynchronous parallel processing (one-step), or do we first figure out what the words of the text mean, and subsequently, and strategically, acquiesce or reject that the new information fits plausibly with our world knowledge. It seems clear that almost every contributor to this volume falls on the “one-step” side of the argument; indeed so markedly, that in the concluding paper, Panayiota Kendeou declares that the theoretical work presented in the volume “in conjunction with evidence already appearing in the literature ….resolves the passive versus strategic controversy in favor of the former” (191). The theorists taken to task on this question are either Gilbert (1991) (who, if I understand correctly, does not maintain that validation occurs in two steps, but that it occurs in one step when people accept a proposition as valid, and that it may involve two steps when people reject a proposition) and/or Connell and Keane (2006) who actually do propose a non-parallel two-step process. 

Isberner and Richter go so far as to say that validation is a “routine component of comprehension, in the sense that comprehension cannot occur independent of validation as both processes rely on the same knowledge activated by memory-based processes” (21), but then have difficulty explaining readers’ vulnerability in the face of false information, particularly that presented in narratives. These researchers conclude that “epistemic monitoring might be suppressed to some extent in narrative (as opposed to argumentative) texts” (23). But wouldn’t this monitoring be instantiated in a two-step process, or if not entirely non-parallel processing, then perhaps in a dual processing model?

A number of factors that may influence validation are discussed in these methodologically tight papers: individual differences in readers (particularly individual standards of coherence); task demands; text characteristics; and source credibility. Some possible factors that it would have been interesting to see discussed, and indeed, if considered, may have guided the editors toward a more balanced representation of the out of favor “two-step” model, are: cross-linguistic effects (This seems somewhat strange, because two of the studies recruited German-as-first-language participants. Besides interesting word order differences as compared to English, German’s tendency toward greater transparency in meaning derivation from a given morphological assembly, for example, might have implications for the validation/ comprehension question. In addition, languages exhibiting evidentiality, in which one must specify the source of one's statement, would also seem a nice test case for linguistic effects on validation.); developmental comparisons (Gilbert relies both on arguments from linguistic evidence and human development in his ostensibly two-step model. All of the participants in the studies reported here are undergraduates.); data from clinical populations (e.g., obsessive/compulsive disorders – the default “doubters”); and the contribution of emotion to validation processing (it's unimaginable that with evidence of emotional states and moods contributing importantly to so many other psychological processes, that it doesn't contribute to validation processing). None of these factors is treated in these papers. 

The paper likely of most direct interest to readers is the one by Rapp, Hinze, Slaten, and Horton. They compared readers’ tendencies to differentially rely on misleading information in unrealistic as compared to fictional narratives (that is fantasy/sci fi versus more mundane domestic narratives). Unrealistic narratives rendered readers less susceptible to believing misleading information, while realistic narratives had the contrary effect. They suggest that “unrealistic settings  encourage accurate sourcing of information to support compartmentalization” (67) and this, combined with the feelings of less fluency with the situations portrayed in science fiction and fantasy combine to impede vulnerability to misinformation. I would suggest reading this paper in the context of the others – an engaging collection.

Cook, A. E., and O’Brien, E.J. (2014). Knowledge activation, integration, and validation during text comprehension. Discourse Processes, 51:1-2, 26-49.

Connell, L., and Keane, M. T. (2006). A model of plausibility. Cognitive Science, 30, 95-120.

Gilbert, D. (1991). How mental systems believe. American Psychologist, 46, 107-119.

Isberner, M.-B., and Richter, T. (2014). Does validation during language comprehension depend on an evaluative mind-set? Discourse Processes, 51: 1-2, 7-25.

Kendou, P. (2014). Validation and comprehension: An integrated overview. Discourse Processes, 51: 1-2, 189-200.

Rapp, D., Hinze, S. R., Slaten, D. G., and Horton, W. S. (2014). Amazing stories: Acquiring and avoiding inaccurate information from fiction. Discourse Processes, 51: 1-2, 50-74.

No comments:

Related Posts Plugin for WordPress, Blogger...