Timothy Desmet’s research while affiliated with Vrije Universiteit Brussel and other places

Little is known about how price evaluation processes unfold. In the current study we explored if reaction times (RTs) can be used to study price evaluations. Additionally, we explored to what extent drift diffusion models (DDMs) are suitable to decompose the different aspects that underlay this decision processes. In a behavioral experiment, participants were asked to evaluate prices as fast as possible as ‘cheap’ or ‘expensive’. We expected that the time needed to evaluate prices would vary in accordance with a price manipulation that was used, and that RTs therefore could be interpreted a proxy of decision difficulty. Analysis of the behavioral data provided evidence for this hypothesis: very cheap and very expensive prices were evaluated faster compared to ambiguous prices. Then, drift diffusion models (DDMs) were used to decompose the different aspect of this decision process, with the goal to obtain a more fine-grained understanding of how the effect in RT data emerged. Results showed that the drift rate of the model was modulated by the price manipulation. Whereas there was no significant effect of the price manipulation on the non-decision time and the starting point parameter. We then contrasted the findings of the RT analysis with the results of the DDMs and outlined what the added value of DDMs is within this context.

The present study employed an explicit reaction time task but measured several underlying cognitive processes in an attempt to provide implicit estimates of consumers’ willingness-to-pay (WTP). Participants were asked to evaluate product-price combinations as cheap or expensive. The prices of the products ranged from very cheap to very expensive. Crucially, participants had to complete the task under time pressure while the dependent variables of interest could not be influenced deliberately. This is because we explored whether the magnitude of the price stimulus interfered with the reaction times (RTs), response force (RF) and partial responses (PRs). The results of our study demonstrated that both RTs and RF are influenced by the magnitude of the price and it is postulated that these dependent measures indeed have the potential to investigate consumers’ WTP. Future studies need to further investigate the possibilities of these implicit variables and validate eventual estimates.

Explicit consumers responses are often adverse for the validity of procedures used to estimate consumers’ willingness-to-pay (WTP). This paper investigates if price evaluations occur automatically and to what extent these automatic processes can be used to implicitly estimate consumers’ WTP. An adapted version of the task-rule congruency (TRC) paradigm was used in two studies. Results of the first study provided evidence for the notion that prices are automatically evaluated. However, the used procedure had limitations that restricted its utility as an implicit WTP estimate. The procedure was adjusted, and an additional study was conducted. The results of the second study also indicated that prices were evaluated automatically. Additionally, the procedure used during the second study allowed to explore to what extent the observed TRC effects could be used to implicitly estimate consumers’ WTP. Taken together, these studies provided evidence for the notion that prices are evaluated automatically. Furthermore, the procedure has the potential to be further developed into an implicit estimate of consumers’ WTP.

In sentences with a complex subject noun phrase, like "The key to the cabinets is lost", the grammatical number of the head noun (key) may be the same or different from the modifier noun phrase (cabinets). When the number is the same, comprehension is usually easier than when it is different. Grammatical number computation may occur while processing the modifier noun (integration phase) or while processing the verb (checking phase). We investigated at which phase number conflict and plausibility of the modifier noun as subject for the verb affect processing, and we imposed a gaze-contingent tone discrimination task in either phase to test whether number computation involves executive control. At both phases, gaze durations were longer when a concurrent tone task was present. Additionally, at the integration phase, gaze durations were longer under number conflict, and this effect was enhanced by the presence of a tone task, whereas no effects of plausibility of the modifier were observed. The finding that the effect of number match was larger under load shows that computation of the grammatical number of the complex noun phrase requires executive control in the integration phase, but not in the checking phase.

Four cross-linguistic structural priming experiments with multilinguals investigated whether syntactic representations for different languages are shared or separate and whether such representations in the first language are stored in a fundamentally different way from those in later acquired languages. The experiments tested whether structural priming within a language differs from priming between languages and whether priming between a first and second language differs from priming between two different second languages. Experiment 1 tested priming of relative clause attachment from Dutch (the subjects’ first language), French, or English (two second languages) to Dutch. Experiments 2 and 3 were similar but had respectively French and English as the target language. Experiment 4 tested dative priming from Dutch, English, and German (another second language) to English. Structural priming was always as strong within- as between-languages and priming between a first and a second language was always as strong as priming between two second languages. These findings support accounts that assume syntax is shared across languages.

We investigated whether the comprehension of syntactically difficult sentences taxes the executive control component of working memory more than the comprehension of their easier counterparts. To that end, we tested the effect of sharing executive control between sentence comprehension and the maintenance of a digit load in two dual-task experiments with strictly controlled timing (Barrouillet, Bernardin, & Camos, 2004). Recall was worse after participants had processed one (Experiment 2) or two (Experiment 1) difficult sentences than after they had processed one or two easy sentences, respectively. This finding suggests that sentence processing and the maintenance of a digit load share executive control. Processing syntactically difficult sentences seems to occupy executive control for a longer time than processing their easy counterparts, thereby blocking refreshments of the memory traces of the digits so that these traces decay more and recall is worse. There was no effect of the size of the digit load on sentence-processing performance (Experiment 2), suggesting that sentence processing completely occupied executive control until processing was complete.

In this paper we show that attachment height (high vs. low attachment) of a modifier to a complex noun phrase (CNP; e.g., "the servant of the actress"), can be primed between dissimilar syntactic structures. In a sentence completion experiment, we found that the attachment height of a prepositional phrase (PP) in the prime sentence primed the attachment height of a relative clause (RC) in the target sentence. This cross-structural priming effect cannot be explained in terms of the priming of specific phrase-structure rules or even sequences of specific phrase-structure rules (Scheepers, 2003), because the attachment of a PP to a CNP is generated by a different phrase-structure rule than the attachment of an RC. However, the present data suggest that the location at which the RC is attached to the CNP is mentally represented, independent of the specific phrase-structure rule that is attached, or by extension, that the abstract hierarchical configuration of the full CNP and the attached RC is represented (Desmet & Declercq, 2006). This is the first demonstration of a cross-structural priming effect that cannot be captured by phrase-structure rules.

A lexical decision experiment with Dutch-English bilinguals compared the effect of word frequency on visual word recognition in the first language with that in the second language. Bilinguals showed a considerably larger frequency effect in their second language, even though corpus frequency was matched across languages. Experiment 2 tested monolingual, native speakers of English on the English materials from Experiment 1. This yielded a frequency effect comparable to that of the bilinguals in Dutch (their L1). These results constrain the way in which existing models of word recognition can be extended to unbalanced bilingualism. In particular, the results are compatible with a theory by which the frequency effect originates from implicit learning. They are also compatible with models that attribute frequency effects to serial search in frequency-ordered bins (Murray & Forster, 2004), if these models are extended with the assumption that scanning speed is language dependent, or that bins are not language specific.

Syntactically ambiguous sentences are sometimes read faster than disambiguated strings. Models of parsing have explained this tendency by appealing either to a race in the construction of alternative structures or to reanalysis. However, it is also possible that readers of ambiguous sentences save time by strategically underspecifying interpretations of ambiguous attachments. In a self-paced reading study, participants viewed sentences with relative clauses that could attach to one of two sites. Type of question was also manipulated between participants in order to test whether goals can influence reading/parsing strategies. The experiment revealed an ambiguity advantage in reading times, but only when participants expected superficial comprehension questions. When participants expected queries about relative clause interpretation, disambiguating regions were inspected with more care, and the ambiguity advantage was attenuated. However, even when participants expected relative clause queries, question-answering times suggested underspecified representations of ambiguous relative clause attachments. The results support the construal and "good-enough" models of parsing.

This article provides an overview of the psycholinguistic research concerning the processes and representations that bilinguals use while processing language. We review the lexical, semantic, and syntactic levels of bilingual processing. The main conclusion from this review is that the different languages of bilinguals strongly influence each other during processing. Therefore, we end with a brief overview of some recent research about how bilinguals are able to keep this interference under control. Throughout the review we will point towards promising future developments.