The role of experience in the interpretation of noun–noun combinations

Gagné and Shoben’s (J Exp Psychol Learn Mem Cogn 23:71–87, 1997) Competition Among Relations In Nominals (CARIN) theory maintains that the interpretation of modifier-noun combinations is influenced primarily by how the modifying noun has been used in the past. As support for this theory, they found that modifiers typically associated with the instantiated relation are interpreted reliably faster than those whose modifiers are less frequently associated with the relation. The CARIN theory explains this phenomenon by proposing that people store statistical distributions regarding the frequency with which modifying nouns have combined with each relation in the past. However, we maintain that an association between relation frequency and response time does not imply a causal influence. In this study we explore whether the effects observed by Gagné and Shoben were caused by the influence of relation frequency per se. Two experiments were conducted in which experiential knowledge about the modifier was controlled. The first experiment involved combinations whose modifiers were relatively rare and the second involved the presentation of nouns without a modifier-head syntax. In both of these experiments, knowledge about historical modifier usage was irrelevant. Our results show that correlations between modifier preference and response time persist even in situations where a knowledge of the modifier’s history is not available. These findings provide converging evidence that the relationship between relation frequency and response time is not a causal one. Instead, an understanding of the relationship between modifier properties and usage, as appropriate to the given context, may be the dominant influence on interpretation in many circumstances. In light of this, we propose an alternative account of the factors influencing ease of interpretation