Over 3 years we engaged with students in the Nidderdale AONB under the auspices of the Wild Watch project. We adapted the games between years to reflect both design deficiences reflected through play and evolving needs on the part of our project partners.
A breakdown of participants follows:
Participant engagement was not found to be age dependent, thus the games are deemed suitable for a variety
of cohorts. The exploration quotient metrics is described in §5.4.3.1
here
While a complete analysis of the first two years' work can be read here (particularly chapters 6 and 7), the following provides a brief summary of both subjective and objective results showing students' learning through play.
Data analysis broadly followed two tracks. The first comprised responses to subjective opinion surveys administered before and after play. The second was founded upon objective analyses of timing and learning error rates for each touch to the devices during each game played; additionally, learning trajectories were examined as participants played multiple games.
In analyzing the data collected from students in the Nidderdale AONB, we found that sound adds value to training for citizen science data collection. The games introduce spectral representations of sound to players who had not previously thought to use both sight and sound for identi cation in a way accessible for students as young as 5.
While prior visual familiarity was consistently higher than audible familiarity, continued audio training though play helps with learning.
As the games were perceived to enhance both visual and audial recognition, we contend that learning success is greatest when a combination of senses is supported.
The data collected from students in the Nidderdale AONB, showed that players were con fident that the games developed visual and audial knowledge across data-representations.
While only short-term knowledge was tested within the scope of our preliminary work in schools, participants were further encouraged to perform a long-term knowledge retention experiment which involves repeating the identification tasks administered pre- and post-play by completing the survey here.
Matching game performance success increased most significantly over the first several iterations of play when participants were exposed to spectral representations of sound.
While purely auditory training requires significantly more repetition, learning to recognize spectrogram
images led to more consistent increases in learning success than with photographs of the calling species,
albeit from a lower baseline.
Students who played the owl games performed pre and post-play owl identi cation tasks; their performance changes provide empirical results of knowledge acquisition.
These were positive for 3 of 4 species taught, indicating that these games have training value. Diff erences between players claimed owl call familiarity and tested call identi cation ability likely result from testing on less common utterances.