Gail McKoon

The Diffusion Decision Model: Theory and Data for Two-Choice Decision Tasks

The diffusion decision model allows detailed explanations of behavior in two-choice discrimination tasks. In this article, the model is reviewed to show how it translates behavioral dataaccuracy, mean response times, and response time distributionsinto components of cognitive processing. Three experiments are used to illustrate experimental manipulations of three components: stimulus difficulty affects the quality of information on which a decision is based; instructions emphasizing either speed or accuracy affect the criterial amounts of information that a subject requires before initiating a response; and the relative proportions of the two stimuli affect biases in drift rate and starting point. The experiments also illustrate the strong constraints that ensure the model is empirically testable and potentially falsifiable. The broad range of applications of the model is also reviewed, including research in the domains of aging and neurophysiology.

Comprehension and recall of text as a function of content variables

Short English texts, controlled for number of words and number of propositions, but differing in the number of word concepts in the text base (many versus few), were read and recalled immediately. Reading times were longer and recall was less for texts with many different word concepts than for texts with fewer word concepts. Superordinate propositions were recalled better than subordinate propositions and forgotten less when recall was delayed. The probability that a word concept was recalled increased as a function of both the number of repetitions of that concept in the text base and the number of repetitions of the corresponding word in the actual text. These results also obtained when subjects listened to the experimental paragraphs.

Inferences about predictable events.

If someone falls off of a 14th story roof, very predictably death will result. The conditions under which readers appear to infer such predictable outcomes were examined with three different retrieval paradigms: immediate recognition test, cued recall, and priming in word recognition. On immediate test, responses to a word representing the implicit outcome (e.g., dead) were slow, but on delayed test these responses were slow or inaccurate only when primed by an explicitly stated word. However, the word expressing the predictable outcome did function as an effective recall cue. Results suggest that readers encode these inferences into memory only minimally, but that they can make use of a cue word that represents the inference (e.g., dead) both at the time of an immediate test and in delayed cued recall.

A diffusion model account of the lexical decision task.

The diffusion model for 2-choice decisions (R. Ratcliff, 1978) was applied to data from lexical decision experiments in which word frequency, proportion of high- versus low-frequency words, and type of nonword were manipulated. The model gave a good account of all of the dependent variables--accuracy, correct and error response times, and their distributions--and provided a description of how the component processes involved in the lexical decision task were affected by experimental variables. All of the variables investigated affected the rate at which information was accumulated from the stimuli--called drift rate in the model. The different drift rates observed for the various classes of stimuli can all be explained by a 2-dimensional signal-detection representation of stimulus information. The authors discuss how this representation and the diffusion models decision process might be integrated with current models of lexical access.

Priming in episodic and semantic memory

Priming between newly learned paired associates was examined in two experimental procedures, lexical decision and item recognition. In lexical decision, the priming effect, shown by decrease in response time, was as large between newly learned associates (e.g., “city grass”) as between semantic associates (e.g., “green grass”). This result shows that episodic information has an effect on semantic (lexical) decisions. In item recognition, priming with semantic associates affected error probability, showing the effect of semantic information on an episodic decision. These results argue against a functional separation of the semantic and episodic memory systems. In the discussion, the utility of the semantic—episodic distinction is examined in some detail.

Spreading activation versus compound cue accounts of priming: mediated priming revisited

Spreading activation theories and compound cue theories have both been proposed as accounts of priming phenomena. According to spreading activation theories, the amount of activation that spreads between a prime and a target should be a function of the number of mediating links between the prime and target in a semantic network and the strengths of those links. The amount of activation should determine the amount of facilitation given by a prime to a target in lexical decision. To predict the amount of facilitation, it is necessary to measure the associative links between prime and target in memory. Free-association production probability has been the variable chosen in previous research for this measurement. However, in 3 experiments, the authors show priming effects that free-association production probabilities cannot easily predict. Instead, they argue that amount of priming depends on the familiarity of the prime and target as a compound, where the compound is formed by the simultaneous presence of the prime and target in short-term memory as a test item.

The comprehension processes and memory structures involved in anaphoric reference.

The inference processes involved in anaphoric reference were examined in three experiments. The first two experiments used an activation procedure in which the subject read a paragraph sentence by sentence and was then presented with a single test word from the first sentence of the paragraph for recognition. Response time to the test word was speeded both when the test word was a referent of an anaphor mentioned in the last sentence of the paragraph and when the test word was in the same proposition as the referent of the anaphor in the last sentence. This shows that an anaphor activates both its referent and concepts in the same proposition as its referent. The third experiment used a priming technique to show that the referent and a concept in the same proposition as the anaphor are connected in the long-term memory representation of a text. These results are discussed in terms of a simple three-process model of anaphoric inference. Other methodologies used to study inference processing are evaluated and it is concluded that these methods involve considerable problems in experimental design and theoretical interpretation and are also limited in the kinds of information they can provide about inference processing.

The activation of antecedent information during the processing of anaphoric reference in reading

The time course of the retrieval of antecedent information during the processing of anaphoric reference was examined in four experiments. In each experiment subjects read paragraphs that appeared one word at a time on a CRT screen. At unexpected times, they were given a single-word recognition test. Response times and error rates for these tests indicated that both the referent (e.g., car) of a superordinate anaphor (e.g., the vehicle) and concepts in the same proposition as the referent become activated as early as 250 milliseconds after the anaphor is read. The referent remains activated as the sentence is read, but the activation of other concepts dies away. The results are interpreted as support for the proposal that antecedent information is initially retrieved in the form of propositions, but only certain concepts from those propositions, those that are important for establishing text structure, remain activated.

A diffusion model analysis of the effects of aging in the lexical-decision task.

The effects of aging on response time (RT) are examined in 2 lexical-decision experiments with young and older subjects (age 60-75). The results show that the older subjects were slower than the young subjects, but more accurate. R. Ratcliff s (1978) diffusion model provided a good account of RTs, their distributions, and response accuracy. The fits show an 80-100-ms slowing of the nondecision components of RT for older subjects relative to young subjects and more conservative decision criterion settings for older subjects than for young subjects. The rates of accumulation of evidence were not significantly different for older compared with young subjects (less than 2% and 5% higher for older subjects relative to young subjects in the 2 experiments).

Priming in Item Recognition: Evidence for the Propositional Structure of Sentences.

A priming technique using item recognition was employed to investigate the structure of the memory representation of simple sentences. The experimental procedure involved presenting sentences to the subject for study and then testing single words for recognition (the subject had to decide whether the test word was in one of the study sentences). A large priming effect was obtained. Response time to a word from one of the study sentences preceded by another word from the same study sentence was over 100 milliseconds faster than response time to a word preceded by a word from a different study sentence. It was hypothesized that, if sentences are stored propositionally, then priming should be greater between words within a proposition than between words in different propositions in the same sentence. This result was obtained and provides strong evidence for the propositional structure of the sentences used in this study. In the discussion, parallels are drawn between priming and the processes used in sentence comprehension, and the process of activation is examined in some detail.