Risk Assessment: The Consumer as an Intuitive Statistician - Innovation and Marketing in the Pharmaceutical Industry - page 300

Knowing someone with a disease can also affect social proximity, but it has not
reliably been shown to reduce self-positivity in estimates of risk. Blalock et al.
( 1990 ) found that individual whose close friend or sibling underwent cancer treat-
ment was more realistic about the likelihood of contracting cancer. Experience may
also lead to perceptions of higher risk for others if the self-positivity bias is due to
people lacking sufficient information about others and failing to consider their cir-
cumstances (Regan et al. 1995 ). Consistent with this argument, Weinstein ( 1980 )
asked students to generate a list of factors that would increase or decrease their
chances of obtaining specific positive and negative future outcomes. A second
group of students who read these lists subsequently became significantly less opti-
mistic about their own chances with respect to negative outcomes.
Weinstein ( 1980 , 1987 ; see also Weinstein and Lachendro 1982 ) used a correla-
tional approach to test the effects of experience on risk perceptions. He used a range
of positive and negative life events (including having a drinking problem and heart
attack) to assess the influence of past experience on self-positivity and found no
correlation between them. Weinstein and Lachendro ( 1982 ) provided detailed, per-
sonalized information about the risk status of five other students and showed a lower
level of the self-positivity bias than others who did not see such information.
Another way in which the self-positivity bias was attenuated in their study was
when the respondents were asked to imagine that they were the typical same-sex
student. Note that in both these experimental conditions, the respondent has not
been provided with an overall base rate, but can bring to mind a specific person or
persons who they are similar to (individuating information), and presumably uses
the information about the individual to visualize that the events happening to the
individual could also occur to them.
Burger and Palmer ( 1992 ) asked undergraduate students who experienced the
1989 California earthquake to estimate their own and others' (average undergradu-
ate at the same school and average person) likelihoods of being hurt in a natural
disaster (e.g., earthquake, flood, and storm). Their results showed that the students
did not show self-positivity when they were asked these estimates within 3 days
after the earthquake, but when they were asked 3 months later, self-positivity
returned. Thus, the effect of experience appears to be contingent on the temporal
proximity of an event.
van der Velde et al. ( 1994 ) asked four different samples that differed in their a
priori actual levels of HIV risk to estimate their risk of HIV infection: general popu-
lation, heterosexual subjects with multiple private partners, gay men with multiple
partners, and visitors to a STD clinic who had engaged in prostitution contact. Only
the group of visitors to the STD clinic who reported having prior experience with an
STD estimated themselves at higher risk than those who did not have such prior
experience and were less prone to self-positivity, possibly reflecting their actual
higher level of risk. All other groups displayed self-positivity.
Probabilistic proximity . Trope et al. ( 2007 ) reason that a low probable event is seen
as more distant than a high probable event, and that increasing probability of an
event lowers the “psychological distance” between oneself and that event. This
implies that one can relate more to high rather than to low probability events.