Transcript cosimo2 - Computer Science Intranet

IS THIS JUST A POSTER DAY?
Supervisor:
Cosimo Nobile
Dr Floriana Grasso
Department of Computer Science
The authors are trying to persuade
the audience about the worthiness
of their research.
If you thought that this
poster day was just part of
the university training
program let me tell you,
from a communication
point of view, there is much
more going on!
The audience might
subsequently need to revise
some beliefs in order to
accept new information.
These two aspects of our every day life are not so
different as they may seem, they can be seen as two
sides, social and cognitive, of the same epistemic coin.
The increasing number of people using the Internet to
support their healthcare, provides new opportunities for
pursuing public health objectives. We look at how to
produce more persuasive pieces of advice to induce
behavioural changes.
Theories such as belief revision (how an individual
changes his own mind), argumentation (how an
individual tries to change other individuals’ mind by
communication), together with other formal rationalchoice models, attempt to mimic how humans reason,
there is of course a discrepancy between
the technicality and constraints of the
theories and the real thing.
DID YOU KNOW?
• A person who buys a Ferrari will likely begin to value
speed over safety in order to avoid dissonance.
• The more you reward someone to tell a lie, the
more public compliance you have but the less
change in private opinion.
• Millennial and messianic movements start
proselytizing more as soon as their prophecies fail.
SO WHAT?
BU
We argue that using CDT as the basis for modelling the
user’s beliefs will bring into play the concept of
motivation and active revision of beliefs and
attitudes for free.
Little attention has been paid to the tendency for
humans to preserve and reinforce our beliefs and
attitudes through time for as long as we can, both for
We also suggest that beliefs and attitudes should be
structured and interconnected to form an “aristocratic
network”, with cognitive dissonance acting as a
major shaping force.
cognitive reasons (revising highly important beliefs, which might also
explain and support many other beliefs, can be extremely costly) and
for sociological reasons (in order to trust our beliefs we pay attention to
very few, and carefully selected, information sources without being too
fussy on reiterations).
COGNITIVE DISSONANCE
According to this theory (CDT), there are
some psychological phenomena that bias
our beliefs, attitudes and behaviour
toward a “dissonance reduction”.
When one behaves in a way that runs counter to one’s
beliefs or when new dissonant information arises, an
aversive motivational state (cognitive dissonance) is
created which pushes the individual towards a belief (or
behaviour) change.
The magnitude of dissonance depends in fact strongly
on the network’s structure: more intricately linked
structures seem more capable of tolerating dissonance
than more simply linked ones. We claim therefore that
an argumentation model that takes into account
such a structure will lead to more effective
computer generated communication.
Computational Model of Behaviour and Belief Change for Persuasive Dialogues
IN OUR CONTEXT…