|
|||||||
|
|
|
|||||
|
|
|||||||
Framing Effects Memory and Illusions Information Processing Prospect Theory Discussion
As mentioned in the section on information processing, information theorists believed memory and reasoning to be very interconnected. So much so, it was difficult to differentiate between them (Reyna & Brainerd, 1991). However, fuzzy-trace theory was an attempt to explain how children and adults apparently do not process certain memories in the course of reasoning. In their research Reyna and Brainerd found that memory and reasoning could be separated, because participant responses differed in precision with varying types of tasks. They proposed that instead of individuals processing information in quantitative numerical terms, individual’s processed information non-quantitatively. They argued that reasoners encode representations at varying levels of precision that can be ordered with respect to precision, and that reasoning gravitates to the lowest, least precise level of gist. What they put forth then can be considered a hierarchy of gist. Gist representation is what stores the semantic content of the material. Recall may be based on an attempt to recover the general theme or gist of material studied. Children and adults search surface forms of inputs for senses and patterns. Items consistent with these patterns will be recalled.
There are different levels of processing that occur in risky decision making problems. Reyna and Brainerd (1995) have equated these levels of processing to the levels of measurement scales introduced by Stevens (Crocker & Algina, 1986). At the ratio scale level, representations preserve precise numerical information. Then at the ordinal level representations capture relative magnitude, but fail to preserve numerical differences (this is the biggest, that’s the next biggest, etc.). Lastly, at the nominal or categorical level only the general presence or absence of quantity is represented (some lives are saved, or no lives are saved). Reyna and Brainerd (1995) argued that reasoning usually prefers to operate on memories that are near the fuzzy end of the continuum (the nominal or categorical level).
Fuzzy-trace
theory makes relatively few assumptions about factors likely to influence
false memories. The first assumption is that memory is not unitary,
but rather that verbal and general surface representation of experience
is encoded roughly in parallel and is stored separately. Although
the meaning of an experience is represented in its general form, that representation
is not integrated with the verbal representation of the same experience
(Reyna & Lloyd, 1997). Subjects may falsely recognize verbal
information because it is elicited memory for the general understanding
of the visual information. According to fuzzy-trace theory, false
memories are generally made worse as time passes between original events
and subsequent presentation of misinformation.
The preference for the level of gist is constrained by the specificity of the response that is required in a given task. Children’s and adult’s memories contain an assortment of reasoning-relevant information that varies in specificity. These range from traces that preserve the exact forms of inputs to traces that preserve only vague information. Choices can be based on the simplest categorical representations, but these are insufficient to provide numerical responses, such as prices. If the response required calculations, then processing would have to occur at a much higher level in the hierarchy of gist. In this fuzzy-trace view, the hierarchy of gist makes task implementations far more robust because the memory functions in a supervisory role rather than a direct implementation mode.
Kuhberger (1995) argued against the gist extraction principle in fuzzy-trace theory. He focused on the amount of information provided to the respondent in framing problems by providing both completely and incompletely described solutions. In his study of the incompletely described version of the Asian disease problem, he found that fuzzy-trace theory yielded no clear predictions, since the two program options used by Reyna & Brainerd (1995) had no identical sections. In their program version of the disease problem, Program A was that some people will not be saved, while program B offered the option that some people will be saved or no one will be saved (Reyna & Brainerd, 1995). They were not clearly logically equivalent options. Kuhberger believed that this level of decision making could not be made at the lowest level on the gist hierarchy (some versus more); it must be made at some higher level. Unlike Reyna and Brainerd, he believed this to be at the numerical level. He argued processing at the numerical level is no longer fuzzy (Kuhberger, 1995).
Fuzzy-trace theory explains age effects by calling attention to young children's greater reliance on verbatim traces, which are less likely to be retrieved and more prone to forgetting and suggestibility in comparison to fuzzy, or gist-like, traces. Verbatim traces deteriorate rapidly and thus may be inaccessible when post-event information is provided or when suggestive questions are asked. Therefore, inaccurate information has an increased likelihood of incorporation with real memories and of becoming indistinguishable from them (Ceci & Bruck, 1993).