Spinozan Hypothesis of Belief

A MORE GENERAL CORRECTION MODEL

The busyness-induced undercorrection of dispositional inferences is now a well-established and widely replicated phenomenon. However, as our work on this problem proceeded, we became convinced that each of our studies was but an instance of a more general phenomenon – namely, the tendency for people to initially believe any notion that they entertained, and to doubt that notion subsequently if at all. If this suggestion sounds odd, it is only because most of us hear it with a Cartesian ear. René Descartes (1644/1984) was the first to suggest that understanding and believing are separate and sequential psychological operations. Three centuries later, the notion that “considering an idea” is different from and prior to “embracing an idea” infuses most modern psychological theorizing. Zimbardo and Lieppe (1991, p. 135) summarized the conventional wisdom as follows:

Learning requires that the audience pay attention to the message and, in turn, gain some comprehension of it, understanding the new beliefs it proposes. Then, if the message has compelling arguments, acceptance of its conclusion and a change in attitude will follow. (italics in original)

What could be more obvious? First, a person comprehends a message, and then later may accept it – which is to say that understanding and believing are the separate and sequential operations that Descartes described.

But Descartes’ contemporary, Benedict Spinoza (1672/1982), did not accept this doctrine, and argued that understanding and believing are a single mental operation. Spinoza suggested that human beings believe assertions as part of understanding them, and that they then “unbelieve” those assertions that are found to be at odds with other established facts. According to Spinoza, people immediately and inevitably accept that which they comprehend, understand, or entertain, and they are incapable of withholding that acceptance. They may change their minds after accepting an assertion, but they cannot stop their minds from being changed by initial contact with it (see Gilbert, 1991, 1992).

Acceptance, then, is a passive and inevitable act, whereas rejection is an active operation that undoes the initial acceptance. The most basic prediction of this Spinozan hypothesis is that when something prevents a person from “undoing” the initial acceptance, then she should continue to believe whatever she is considering – even when what she is considering is patently false. For example, if a person is told that lead pencils are a health hazard, the Spinozan hypothesis suggests that she must immediately believe that assertion and only then take active measures to unbelieve it. These active measures may require cognitive work (i.e., the search for or generation of contravening evidence), and if something impairs the person’s ability to perform such work, then the person should continue to believe in the danger of lead pencils until the cognitive work can be done. The Cartesian hypothesis, however, makes no such predictions. That hypothesis suggests that both acceptance and rejection of an assertion are the results of cognitive work that follows comprehension of the assertion. As such, a shortage of resources should make both of these options impossible, thus leaving the person in a state of non-belief, rather than a state of belief or disbelief.

In a series of studies (Gilbert, Krull, & Malone, 1990), we attempted to provide direct evidence for the Spinozan hypothesis by presenting participants with assertions whose veracity they could not assess because one word of the assertion was in a foreign language. Participants read 28 propositions, one at a time, of the form An X is a Y, where X was ostensibly a Hopi noun and Y was its ostensible English equivalent. Participants were told that at the end of the session their knowledge of the Hopi vocabulary would be tested. After seeing each proposition on the computer screen, the participants saw the word true or false, which indicated that the preceding proposition had been either accurate or inaccurate. In addition, participants occasionally heard a tone during the experiment, and when this happened they were instructed to press a response button as quickly as possible. This tone-detection task was designed to interrupt participants’ processing of the proposition that was currently being presented on the computer screen.

The Spinozan hypothesis (but not the Cartesian hypothesis) predicted that interruption would (1) prevent participants from unbelieving the assertions that they automatically accepted on comprehension and would thus cause participants to report that false assertions were true, but (2) would not cause participants to report that true assertions were false. As Table 9.7 shows, precisely this asymmetry emerged – an asymmetry that is not easily explained by the Cartesian hypothesis. The Spinozan hypothesis was also unique in suggesting that the processing of a false proposition should require more time and cognitive capacity than the processing of a true proposition, simply because false propositions must be accepted and then unaccepted (see also Carpenter & Just, 1975; Clark & Chase, 1972, 1974; Gough, 1965, 1966; Trabasso, Rollins, & Shaughnessey, 1971). An analysis of the time it took participants to respond to the interrupting tone revealed that they responded significantly more slowly to interruptions that occurred immediately after the presentation of a false proposition than to those that occurred immediately after the presentation of a true proposition. Supplementary data and studies ruled out the possibilities that interruption merely kept participants from remembering false propositions, or led them to feel uncertain about the veracity of propositions and thus led them to guess that the items were true.