Concluding Remarks and Pointers - Trust Theory: A Socio-Cognitive and Computational Model - page 353

13

Concluding Remarks and Pointers

Let us conclude this quite complex topic, on the one hand, with some brief considerations,
on the other hand, with two short paragraphs pointing to two very relevant and very topical
developments: the new frontiers of neuro-science of trust; political dimensions of trust and of
its crisis.

13.1 Against Reductionism
As for the general remarks, we hope that our readers have realized that trust is not at all
a simple object and notion, but that it is possible and necessary (or at least very useful) to
recollect its different aspects, functions, and components in a unitary, integrated, principled,
and well-structured frame. This has been our attempt; against a lot of very reductive treatments
of trust, due to very limited and ad hoc domains or disciplinary interests, or to very rigid formal
tools.
We have systematically argued against not just abstraction, simplification, or normativity
(very necessary in science), but against deformation and reductionism: arguing, for example,
against the reduction of trust to a mere mental attitude without a behavioral or relational
component, or to a mere vague feeling without specific thoughts (evaluations, predictions, and
so on), or to a mere measure of the frequency of successes, or to mere subjective probability of
a favorable event, or just to behavioral reliance, or just to exchange situations and expectation
of reciprocation, and so on.
Moreover, it is not at all a question of taking into account all the components and functions,
just adding one to the other, but of showing how one property is based on the other, one
component is integrated with the others in a precise structure.
Finally, the structural nature of trust should not just be its 'statics', but it should also find and
justify its 'dynamics': how trust increases or decreases, how we build it, how trust propagates,
or how it is generalized, instantiated, transferred from one agent or one task/good to another,
and so on.
No doubt our solution is still unsatisfactory, but we hope that our readers will be persuaded
that this is the right objective: an explicit anatomy of trust, an integrated and justified model of
its ingredients, of their integration, and of how it works . We do not need just domain specific