Facilitation and inhibition caused by the orienting of attention in propositional reasoning tasks

Abstract

In an attempt to study the orienting of attention in reasoning, we developed a set of propositional reasoning tasks structurally similar to Posner's (1980) spatial cueing paradigm, widely used to study the orienting of attention in perceptual tasks. We cued the representation in working memory of a reasoning premise, observing whether inferences drawn using that premise or a different, uncued one were facilitated, hindered, or unaffected. The results of Experiments 1a, 1b, 1c, and 1d, using semantically (1a–1c) or statistically (1d) informative cues, showed a robust, long-lasting facilitation for drawing inferences from the cued rule. In Experiment 2, using uninformative cues, inferences from the cued rule were facilitated with a short stimulus onset asynchrony (SOA), whereas they were delayed when the SOA was longer, an effect that is similar to the “inhibition of return” (IOR) in perceptual tasks. Experiment 3 used uninformative cues, three different SOAs, and inferential rules with disjunctive antecedents, replicating the IOR-like effect with the long SOAs and, at the short SOA, finding evidence of a gradient-like behaviour of the facilitation effect. Our findings show qualitative similarities to some effects typically observed in the orienting of visual attention, although the tasks did not involve spatial orienting.

The authors thank P. Legrenzi, C. Umiltà, B. Scholl, R. Dell'Acqua, E. Bricolo, B. Hommel, B. Burns, and five unknown referees to previous versions of this paper for their helpful comments and A. Zanella and C. Garbin for their help in gathering the data.

Notes

¹ Neutral cues are obtained in many ways, none of them entirely satisfying (Jonides & Mack, 1984).

² Symbolic/central cues cause attentional orienting only if they are informative (but see Hommel, Pratt, Colzato, & Godijn, 2001, for an exception). Peripheral cues—that is, those that signal a location by their abrupt onset or offset (Pratt & McAuliffe, 2002; Riggio, Scaramuzza, & Umiltà, 2000)—cause reflexive facilitation even if they are uninformative.

³ Read “ → ” as “if … then … ”, “&” as “and”.

⁴ The “matching bias” in reasoning (Evans, 1972, 1996; Evans & Lynch, 1973) originates from a tendency to focus attention on data that match the components of a conditional rule; it is a plausible conjecture that also the complement to it might be true: That is, a piece of information matching one of two rules might focus attention on that rule.

⁵ Given a conditional sentence, “if p then q”, and a categorical premise affirming its antecedent, “p”, the modus ponens inference is the affirmation of the consequent, “q”. Modus ponens inferences are the simplest inferences in the human repertoire, and they are almost error free under normal conditions (see Evans, Newstead, & Byrne, 1993, for a review of the main studies on availability of propositional inferences).

⁶ Many studies have shown that one system for the orienting of visual attention is object centred: That is, it selects whole visual objects, and is advantaged when processing one object at a time. For example, comparing two identical details is faster when the two details pertain to the same object and slower when the two details are part of two different objects, everything else being kept the same (e.g., Baylis, 1994; Behrmann, Zemel, & Mozer, 1998; Duncan, 1984; Egly, Driver, & Rafal, 1994; Mapelli, Cherubini, & Umiltà, 2002; Vecera & Farah, 1994; Watson & Kramer, 1999).

⁷ To give some examples: The “dual process” theory of reasoning by Evans and colleagues (Evans, 1984, 1996; Evans & Over, 1996) holds that the first step of reasoning is a preattentive filtering-out of presumably irrelevant information. This preattentive filter helps explaining bad performance in Wason's selection task (Evans, 1998; Evans & Lynch, 1973). Likewise, built-in filters of presumably irrelevant information are described by mental models theorists: The “focusing effect” (Legrenzi, Girotto, & Johnson-Laird, 1993) and the “illusory inferences” (Johnson-Laird & Savary, 1999) are probably the result of a filtering mechanism (termed the “principle of truth”).

⁸ IOR is present in perceptual discrimination tasks (Chasteen & Pratt, 1999; Lupiáñez, Milán, Tornay, Madrid, & Tudela, 1997; Pratt, 1995; Pratt & Abrams, 1999; Terry, Valdes, & Neill, 1994).

⁹ From a syntactic, rule-based perspective one should assume that the movement of attention between two elements of the premises follows a tree-like, hierarchic path determined by how the premises are nested within the problem:

Given this assumption, figure₁ and figure₃ are further apart (intuitively you must cross many parentheses to get to figure₃ starting from figure₁) than figure₁ and figure₂ (no parentheses to be crossed). Likewise, from a model-based perspective, one should assume that recursive leaps in the hierarchy of models increase the distances between their elements. Namely, given that the “molecular” model of disjunctive₁ comprises the “atomic” models of figure₁ and of figure₂, whilst the model of disjunctive₂ comprises the “atomic” models of figure₃ and figure₄, the distance from figure₁ to figure₃ should be greater than that from figure₁ to figure₂, because the former requires going from a “molecular” model to a different one, whereas the latter remains within the same “molecule”. To our knowledge, neither mental models theory nor any rule-based approach has ever made similar assumptions (probably because they never dealt formally with the problem of how attention is shifted between the different premises of a problem), but in our opinion both of them would retain their consistency if they did.