Academic journal article Memory & Cognition

Is Inferential Reasoning Just Probabilistic Reasoning in Disguise?

Academic journal article Memory & Cognition

Is Inferential Reasoning Just Probabilistic Reasoning in Disguise?

Article excerpt

Oaksford, Chater, and Larkin (2000) have suggested that people actually use everyday probabilistic reasoning when making deductive inferences. In two studies, we explicitly compared probabilistic and deductive reasoning with identical if-then conditional premises with concrete content. In the first, adults were given causal premises with one strongly associated antecedent and were asked to make standard deductive inferences or to judge the probabilities of conclusions. In the second, reasoners were given scenarios presenting a causal relation with zero to three potential alternative antecedents. The participants responded to each set of problems under both deductive and probabilistic instructions. The results show that deductive and probabilistic inferences are not isomorphic. Probabilistic inferences can model deductive responses only using a limited, very high threshold model, which is equivalent to a simple retrieval model. These results provide a clearer understanding of the relations between probabilistic and deductive inferences and the limitations of trying to consider these two forms of inference as having a single underlying process.

Conditional (if-then) reasoning is considered by many as the cornerstone of deductive reasoning and has been the subject of an important amount of research. Although deductive reasoning is ideally considered to be abstract in nature, there are many striking forms of variation, due to premise content, in the way that even educated adults make deductive inferences. These suggest that the processes that underlie deduction are sensitive to knowledge about the premises in some way. This form of variation has led Oaksford, Chater, and Larkin (2000) to propose that when people make conditional inferences, they do not attempt to make "logical" inferences but, rather, use everyday probabilistic reasoning, which directly accesses knowledge about premises, to respond to these problems. However, although such a model can account for variation in performance in many instances, there are other ways of doing this that retain the distinction between probabilistic and inferential processes. In the following, we will present two studies that compare deductive and probabilistic reasoning on the same problems.

Conditional reasoning consists in making an inference on the basis of a major premise of the type "if P, then Q" (the first clause, P, is called the antecedent; the second, Q, is called the consequent). There are four basic inferences that can be made from a major conditional premise. Two of these inferences, the affirmation of the antecedent ("P is true"), called modus ponens (MP), and the denial of the consequent ("Q is false"), called modus tollens (MT), lead to logically certain conclusions. For example, assuming the truth of the proposition "if it rains, then the street is wet" authorizes the conclusions that "if it rains," then it is certain that "the street is wet" and "if the street is not wet," then "it is not raining." The two other logical forms, consisting of the denial of the antecedent (DA; "P is false") and the affirmation of the consequent (AC', "Q is true") are uncertain arguments, because they do not allow logically certain conclusions. For example, the DA "it is not raining" does not permit the certain conclusion that "the street is not wet," because it is possible that something other than rain can cause the street to be wet. Likewise, knowing that "the street is wet" (AC) does not permit the certain conclusion that "it has rained," for the same reason.

Research has consistently shown that adult reasoners will often draw what are logically inappropriate conclusions when asked to make deductive inferences. This is particularly the case with the two logical forms for which there is no certain conclusion, AC and DA, where a common response is to accept the invited inference in each case. Several studies have shown that one key factor that appears to influence what kinds of inferences are made on the AC and DA forms concerns the relative number and availability of alternative antecedents-that is, cases of [not-P and Q] (Cummins, 1995; Cummins, Lubart, Alksnis, & Rist, 1991; Janveau-Brennan & Markovits, 1999; Markovits & Vachon, 1990; Quinn & Markovits, 1998; Thompson, 1994). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.