Socio-Cognitive Model of Trust: Quantitative Aspects - Trust Theory: A Socio-Cognitive and Computational Model - page 104

success

10

to do by itself

1

failure

success

50

delegation

5

failure

Figure 3.7 An Example in which the maximum acceptable damage is over the threshold. (Reproduced
with kind permission of Springer Science+Business Media C

2001)

It is possible that all the branches in the decision scenario would be in a situation of
saturation (

1). What choice does the agent have to make? In these cases there could be
several different possibilities.
Let us consider the scenario in Figure 3.5. There could be at least four possibilities:

σ H =

1) Saturation due to σ d for branch 'to do by itself' (the potential damage is too high); saturation

σ a for branch 'delegation' (the payoff is too low).
2) Saturation due to

due to

σ a for branch 'to do by itself' (the payoff is too low); saturation due to
σ d for branch 'delegation' (the potential damage is too high).
3) Saturation due to

σ a for branch 'to do by itself' (the payoff is too low); saturation due to
σ a for branch 'delegation' (the payoff is too low).
4) Saturation due to σ d for branch 'to do by itself' (the potential damage is too high); saturation

due to σ d for branch 'delegation' (the potential damage is too high).

In the cases (1) and (2) the choice could be the minimum damage (better a minor payoff
than a high damage).
In the case (3) if ( σ a −

σ a − U(X) p + ) then the choice will be 'to do by itself'
and vice versa in the opposite case. In other words, given a payoff threshold greater than
both the utilities (performance and delegation) the action with greater utility (although
insufficient) will be better.

U(X) d + ) < (

success

20

to do by itself

10

failure

success

15

delegation

4

failure

Figure 3.8 An Example in which the minimal acceptable value is under the threshold