Dynamics of Trust - Trust Theory: A Socio-Cognitive and Computational Model - page 167

With X, Z, . ., Z n Y ∈ S and σ 1 , .., σ n respectively their thresholds for delegating.

A very crucial example of this trust-spreading mechanism is about circulating information.
The greater the number of (independent) sources of a piece of information, a reputation, a
belief, the more credible it is. (Castelfranchi, 1996)
In fact, belief acceptance is not an automatic process, it is subject to some tests and checks:
one check is for its plausibility, coherence with previous knowledge, for the source reliability,
etc. Thus, if several cognitive agents believe bel 1 , probably bel 1 has been checked by each
of them against its own direct experience, previous knowledge, and source evaluation, thus it
is reasonable that it is more credible. Even in science, convergence is a criterion of validity.
However, this is also a rather dangerous mechanism (of social bias and prejudice) since in fact
sources in a social context are not independent and we cannot ascertain their independence,
thus the number of sources can be just an illusion: there could be just a unique original source
and a number of 'reporters'.
All these forms of trust spreading are converging in a given target of trust (being it an
information source, a technology, a social practice, a company, a guy).

6.5.2 Since X Trusts Y , (by Analogy) Z Trusts W
This is a generalized form of trust contagion. It can be based on cognitive analogy or co-
categorization:

since X trusts Y (and X is expert) and W is like Y , Z will trust W
or
since everybody trusts some Y , and W is like/a Y , Z will trust W .

Where 'like' either means ' W is similar to Y as for the relevant qualities/requirements', or
means ' W is of the same category of Y '; and 'some Y ' means 'someone of the same category
of Y or similar to Y '. Also in this case either a specific model and example counts, or a
more generalized practice. For a more diffused and analytic treatment of this argument see the
following Section 6.6.

6.5.3 Calculated Influence
As we said, usually, this trust contagion is a macro-level phenomenon and the individual agents
do not calculate it, it is not intended. However, sometimes it is a strategic decision. Suppose
for example that X believes herself to be a 'model' for Z , and that she wants Z to trust Y (or
W ). In this case X can deliberately make Z believe that he trusts Y , in order to influence Z and
induce him to trust Y (or W ).
The smart businessman might be aware of the fact that when he buys certain shares, a lot
of other people will follow him, and he can exploit precisely this imitative behavior of his
followers and speculate at their expense.