Over-specified referring expressions impair comprehension: An ERP study

• Published in: Brain and cognition Vol. 77; no. 2; pp. 304 - 314
• Main Authors: Engelhardt, Paul E., Barış Demiral, Ş., Ferreira, Fernanda
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01-11-2011; Elsevier
• Subjects: Adjective; Adolescent; Adult; Attention; Attention - physiology; Behavioral psychophysiology; Biological and medical sciences; Brain Hemisphere Functions; Cerebral Cortex - physiology; Cognitive Processes; Comprehension - physiology; Cues; Diagnostic Tests; Electroencephalography; Electrophysiology; Evoked Potentials - physiology; Female; Form Classes (Languages); Fundamental and applied biological sciences. Psychology; Humans; Internet; Language; Language Processing; Male; Maxim of Quantity; N400; Over-description; Photic Stimulation; Pre-nominal modifier; Production and perception of spoken language; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Reaction Time; Reaction Time - physiology; Semantic integration; N400; Semantic integration; Adjective; Over-description; Pre-nominal modifier; Maxim of Quantity; Human; Verbal comprehension; Central nervous system; Electrophysiology; Cognition; Encephalon; Language; Semantics; Speech; Event evoked potential
• ISSN: 0278-2626; 1090-2147
• DOI: 10.1016/j.bandc.2011.07.004

► We examined whether comprehension is impaired with too much information. ► RTs and ERPs revealed that an extra modifier slowed comprehension and produced an N400. ► Data are consistent with the Maxim of Quantity.
Speakers often include extra information when producing referring expressions, which is inconsistent with the Maxim of Quantity (
Grice, 1975). In this study, we investigated how comprehension is affected by unnecessary information. The literature is mixed: some studies have found that extra information facilitates comprehension and others reported impairments. We used an attentional-cueing paradigm to assess how quickly participants could orient attention to an object upon hearing a referring expression, such as
the red square. If there are two squares differing in color, then the modifier is required. However, if there is only one (red) square, then the modifier is unnecessary. We also recorded event-related potentials (ERPs) in order to investigate online processing. Reaction times were significantly longer for referring expressions that contained extra information, and ERPs revealed a centroparietal negativity (N400) that emerged approximately 200–300
ms after modifier onset. We conclude that referring expressions with an unnecessary pre-nominal modifier impair comprehension performance.