Game Development Reference
In-Depth Information
Table 2. Mapping between numbers and outcome-
prospective emotions
emotion will be displayed on it. In Figure 8, the
player-character's suit is white, which signifies
no-emotion.
PlayPhysics ' teaching effectiveness will be
evaluated by separating students enrolled in
introductory Physics courses into control and
experimental groups. Both groups will perform a
pre-test to determine their knowledge and under-
standing on Physics topics before receiving the
corresponding teaching. The experimental group
will interact with PlayPhysics after receiving the
corresponding lectures. At the end both groups will
do a post-test. The learning gains will be compared
and the learning efficiency will be calculated. The
latter is obtained through the quotient of the times
that the student succeeded in the challenge and
the total time that the student played.
Number
Emotion
1
Anticipatory relief
2
Anxiety
3
Hopelessness
4
Hope
5
Anticipatory joy
6
No emotion
dents. In addition, the main aim of this preliminary
test was to identify deficiencies in the prototyping
materials. Some postgraduate students reported
that the pre-test contained Physics definitions as
formulae and they did not know what the variables
in the formulae meant and suggested that an ex-
planation of the terms would make it easier to
reason about them. Other postgraduate students
reported that the pre-test would be easier to reason
about if a diagram was included to support the
question. In questions related to the game dialogue,
it was signalled that the option 'no-emotion' must
be added to the set of emotions that the student
can report. Also, it was noticed that some student's
did not know how to report the emotion they were
feeling, e.g. one student reported that he felt
anxious, but not in a negative way. On the other
hand, in two out of three specific cases where the
students reported 'anxiety' instead of 'anticipa-
RESULTS AND DISCUSSION
Seven participants, postgraduate students study-
ing at Ph.D. level, answered a pre-test, which
comprised 5 questions related to the topics in
PlayPhysics ' first challenge. After completing the
pre-test and being made aware of their results, they
answered the game dialogue questions according
to their attitudes, beliefs and previous perfor-
mance experience. The participants also reported
the outcome-prospective emotion that they felt
before starting the first challenge. The acquired
evidence of the pre-test and of the answers to the
game-dialogue was propagated in the outcome-
prospective emotions DBN. Results, reported in
Table 2 and Table 3 and the graph in Figure 9,
show that in this test the affective student model
had an accuracy of 57.14%, i.e. out of 7 attempts
the emotion was accurately inferred four times.
The nature of these results may be due to the
probabilities set in the Conditional Probability
Tables (CPTs) using common sense, and due to
the small population of participants, since to
validate the Achievement Emotions Questionnaire
(AEQ), Pekrun et al. (2005) employed 389 stu-
Table 3. Emotion reported by the student vs. emo-
tion predicted by the system
Number of
student
Emotion reported by
the student
Emotion predicted
by the system
1
2
2
2
5
5
3
2
5
4
2
5
5
2
5
6
4
4
7
5
5
 
Search Nedrilad ::




Custom Search