Skip to content

Heuristics and Biases 2B- Support Theory


25. Support Theory: A Nonextensional Representation of Subjective Probability

For the most part, this chapter explores the fact that explicitly stating hypotheses will increase their rated probability. For example, asking what the probability of dying from “natural causes” is will be rated less than asking what the probability of dying from a heart attack, stroke, or any other natural cause. The second comprises all “natural causes” so it should be equal, but explicitly stating these hypotheses increases their rating. When the hypotheses are implicit, they are “subadditive” i.e. they do not add up to enough to get up to the level of the entire class of hypotheses.

I could see this idea getting hacked in polling. If you broke up Christianity into 5 different types, but did not do so for other religions, than asking someone the probability that each “religion” is true would lead to a buffed Christianity stat. It also shows why certain scientific hypotheses are discounted. If you ask the probability that an all-powerful God created the universe, vs. the probability of “all other explanations” there will be some neglect of exactly how many explanations fall into the “all other” category, and they will be rated lower in probability than if you just spelled them out.

The degree of subadditivity increases with the number of components in the explicit disjunction. This follows readily from support theory: Unpacking an implicit hypothesis into exclusive components increases its total judged probability, and additional unpacking of each component should further increase the total probability assigned to the initial hypothesis (10460).

The empirical evidence confirms the major predictions of support theory: (1) Probability judgments increase by unpacking the focal hypothesis and decrease by unpacking the alternative hypothesis; (2) subjective probabilities are complementary in the binary case and subadditive in the general case; and (3) subadditivity is more pronounced for probability than for frequency judgments, and it is enhanced by compatible evidence (10700).

26. Unpacking, Repacking, and Anchoring: Advances in Support Theory

Sometimes even explicitly unpacked hypotheses are underweighted, and suffer from subadditivity. This is because of repacking, the act of taking multiple hypotheses and putting them back into a packed one, and anchoring and adjusting. An example of this would be estimating the population of the U.S., Canada, and Mexico combined. One might anchor on the U.S. (~300 mil), and adjust upwards to account for Canada and Mexico, instead of estimating each country. Usually this adjustment is insufficient.

We propose that implicit subadditivity is caused by enhanced availability, whereas explicit subadditivity is produced, in part at least, by repacking or anchoring (11029).

27. Remarks on Support Theory: Recent Advances and Future Directions

This chapter applies support theory to other well known heuristics/biases. The idea of “support” accounts for a multitude of observations, like the self-serving bias (above average always), and confirmation bias.

Support theory, a formal descriptive account of subjective probability introduced by Tversky and Koehler (1994), offers the opportunity to weave together the different heuristics into a unified account. The theory can accommodate many mechanisms (such as the various heuristics) that influence subjective probability, but integrates them via the construct of support. Consequently, support theory can account for numerous existing empirical patterns in the literature on judgment under uncertainty (11099).

Gilovich (2000) suggests that conclusions a person does not want to believe are held to a higher standard than conclusions a person wants to believe. In the former case, the person asks if the evidence compels one to accept the conclusion, whereas in the latter case, the person asks instead if the evidence allows one to accept the conclusion (11441).

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: