trustee to accomplish a task, to give correct information and so on. A specific trust
subgoal may involve both constructs to some degree or may depend on a single
one for the most part. The distinction between these constructs impacts trust models
significantly. There are likely many differences in what signals are used to evaluate
them, how they impact decisions and how firsthand evidence is incorporated into
trust beliefs. Plus, these two beliefs are distinct: a trustee may be considered both
trustworthy and competent, but it is also possible that the trustee is only considered
either trustworthy or competent.
Trustworthiness is highly correlated with friendship, in-group characterizations,
and having the same values. There is some evidence that negative evidence carries
more weight. If our friend is acting trustworthy, this is not significant. Hence,
the trustworthiness belief is not impacted very highly with positive evidence.
However, if he does something wrong, the trustworthiness belief may change
drastically. As with many cognitive constructs, strong inputs, i.e., very positive or
negative experiences, are likely to make an impact on the trustworthiness beliefs.
Trustworthiness beliefs have an affective component, e.g., relying on someone to
come through for you, and are not dependent on a topic or a domain of expertise.
Faces play a major role in judging how trustworthy someone is, and are
evaluated very quickly and with little effort. This, trustworthiness evaluations made
through the assessment of faces are available before any other trust evaluation.
Such first impressions can be eventually overwritten by information obtained
through second-hand knowledge (stories, reviews, reputation ratings) and firsthand
experiences (true social relations). However, research shows that first impressions
have a significant impact and may impact how the new evidence for a trustee is
Having common values and shared intentionality is a major component of
judging someone to be trustworthy. In fact, this is cited as the main reason that
automated systems cannot be trusted in the same way as human beings. If the
trustee is considered not to share the trustor's cultural values, then he is likely
to be distrusted. This means that other factors do not matter. Furthermore, such
a determination depends heavily on the level of risk. For a decision requiring a
high level of risk, the trustor is likely to choose a trustee that she finds trustworthy,
regardless of their competence.
An analogue of trustworthiness exists when evaluating information.
The visual appeal and organization of a site offering information provides cues about
how professional and reliable the information is. As with faces, these evaluations
are made quickly and without much effort. Most people form a judgment about
a site in a fraction of a second. This impacts which sites will be visited. If a site
is not visited, it will not get a chance to be evaluated for trustworthiness and/or
competence. This problem is similar to our discussion in Sect. 5.1 .
There is also evidence that competence is a different type of belief. For example,
positive evidence is considered more relevant for judging competence. The more
evidence that demonstrates Bob's knowledge in a field, the more likely Alice will
believe that he is an expert in that field. Alice may then be willing to forgive