The Effect of Prominence and Cue Association on Retrieval Processes: A Computational Account

Abstract

We present a comprehensive empirical evaluation of the ACT-R-based model of sentence processing developed by Lewis and Vasishth (2005) (LV05). The predictions of the model are compared with the results of a recent meta-analysis of published reading studies on retrieval interference in reflexive-/reciprocal-antecedent and subject-verb dependencies (J€ ager, Engelmann, & Vasishth, 2017). The comparison shows that the model has only partial success in explaining the data; and we propose that its prediction space is restricted by oversimplifying assumptions. We then implement a revised model that takes into account differences between individual experimental designs in terms of the prominence of the target and the distractor in memory-and context-dependent cue-feature associations. The predictions of the original and the revised model are quantitatively compared with the results of the meta-analysis. Our simulations show that, compared to the original LV05 model, the revised model accounts for the data better. The results suggest that effects of prominence and variable cue-feature associations need to be considered in the interpretation of existing empirical results and in the design and planning of future experiments. With regard to retrieval interference in sentence processing and to the broader field of psycholinguistic studies, we conclude that well-specified models in tandem with high-powered experiments are needed in order to uncover the underlying cognitive processes.

The effect of decay and lexical uncertainty on processing long-distance dependencies in reading

“…The evidence against an effect of temporal decay in both the self-paced reading and eye tracking experiments is consistent with findings suggesting that decay is not an important factor influencing reading and memory recall times (Lewandowsky, Oberauer & Brown, 2009;Engelmann, Jäger & Vasishth, 2019;Vasishth et al, 2019). In comparison to the sentences used in distance manipulations in previous studies, our sentences used simple adjectival modifiers that deliberately avoided the introduction of interference or new discourse referents.…”

“…While activation decay may be a factor in sentence processing, there is evidence to suggest that it is not a useful predictor of processing difficulty ( Van Dyke & Johns, 2012 ; Engelmann, Jäger & Vasishth, 2019 ; Vasishth et al, 2019 ), and that longer word recall times and reduced accuracy over time are better explained by interference than decay ( Lewandowsky, Oberauer & Brown, 2009 ). On the other hand, much of this evidence comes from computational modelling based largely on data from experiments testing interference rather than specifically testing decay.…”

A Principled Approach to Feature Selection in Models of Sentence Processing

“…Research in human sentence comprehension and production increasingly relies on computational models that implement competing psycholinguistic theories to generate testable predictions about human behavior. These models take a variety of forms, including models based on cue‐based memory retrieval (Engelmann, Jäger, & Vasishth, 2019; Lewis & Vasishth, 2005; Vasishth, Nicenboim, Engelmann, & Burchert, 2019), self‐organization (Smith, Franck, & Taborr, 2018; Smith & Tabor, 2018; Tabor & Hutchins, 2004), and expectation‐based parsing (Futrell & Levy, 2017; Hale, 2001; Levy, 2008). They aim to explain well‐established sources of processing difficulty like garden paths (Bever, 1970), local coherence effects (where a string of words is locally well‐formed but ungrammatical in the context of the rest of the sentence; Tabor, Galantucci, & Richardson, 2004), and similarity‐based interference (where the presence of words with similar features in a sentence affects processing; Jäger, Engelmann, & Vasishth, 2017).…”

“…The self‐organizing theory in Smith et al (2018), however, relies solely on semantic features to bias the competition for which noun controls the verb number. Finally, the cue‐based retrieval model of Lewis and Vasishth (2005), recently extended in Engelmann et al (2019), typically uses a structural feature which refers to a tree‐configurational property (such as +subject or +c‐command) in combination with another morphological or semantic feature (such as +singular or +animate) to determine the speed and probability of retrieving a noun to set the verb number (see below for more details).…”

Table 7: Experiment 2: Summary of question response accuracy and reaction times.

“…While activation decay may be a factor in sentence processing, there is evidence to suggest that it is not a useful predictor of processing difficulty (Van Dyke & Johns, 2012;Engelmann, Jäger & Vasishth, 2019;Vasishth et al, 2019), and that longer word recall times and reduced accuracy over time are better explained by interference than decay (Lewandowsky, Oberauer & Brown, 2009). On the other hand, much of this evidence comes from computational modelling based largely on data from experiments testing interference rather than specifically testing decay.…”