Limits...
Modeling strategic use of human computer interfaces with novel hidden Markov models.

Mariano LJ, Poore JC, Krum DM, Schwartz JL, Coskren WD, Jones EM - Front Psychol (2015)

Bottom Line: We further report the results of a preliminary study designed to establish the validity of our modeling approach.Pre- and post-task questionnaires probed for self-reported styles of problem solving, as well as task engagement, difficulty, and workload.Overall, we find that this novel approach to decomposing unstructured behavioral data within software environments provides a sensible means for understanding how users learn to integrate software functionality for strategic task pursuit.

View Article: PubMed Central - PubMed

Affiliation: The Charles Stark Draper Laboratory, Inc. Cambridge, MA, USA.

ABSTRACT
Immersive software tools are virtual environments designed to give their users an augmented view of real-world data and ways of manipulating that data. As virtual environments, every action users make while interacting with these tools can be carefully logged, as can the state of the software and the information it presents to the user, giving these actions context. This data provides a high-resolution lens through which dynamic cognitive and behavioral processes can be viewed. In this report, we describe new methods for the analysis and interpretation of such data, utilizing a novel implementation of the Beta Process Hidden Markov Model (BP-HMM) for analysis of software activity logs. We further report the results of a preliminary study designed to establish the validity of our modeling approach. A group of 20 participants were asked to play a simple computer game, instrumented to log every interaction with the interface. Participants had no previous experience with the game's functionality or rules, so the activity logs collected during their naïve interactions capture patterns of exploratory behavior and skill acquisition as they attempted to learn the rules of the game. Pre- and post-task questionnaires probed for self-reported styles of problem solving, as well as task engagement, difficulty, and workload. We jointly modeled the activity log sequences collected from all participants using the BP-HMM approach, identifying a global library of activity patterns representative of the collective behavior of all the participants. Analyses show systematic relationships between both pre- and post-task questionnaires, self-reported approaches to analytic problem solving, and metrics extracted from the BP-HMM decomposition. Overall, we find that this novel approach to decomposing unstructured behavioral data within software environments provides a sensible means for understanding how users learn to integrate software functionality for strategic task pursuit.

No MeSH data available.


Example of a discrete observation probability distribution of the 60-symbol alphabet used to encode the activity logs and temporal features. Each small bar corresponds to a centroid representing a range of temporal inter-activity values. The height of the bar represents the relative probability of observing the corresponding symbol. The bars are grouped by activity log type, and the height of the larger blocks is equal to the total probability of observing each activity log, summing across all temporal values. The actual inter-activity times represented by one symbol from each group are depicted below the graph.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4490801&req=5

Figure 4: Example of a discrete observation probability distribution of the 60-symbol alphabet used to encode the activity logs and temporal features. Each small bar corresponds to a centroid representing a range of temporal inter-activity values. The height of the bar represents the relative probability of observing the corresponding symbol. The bars are grouped by activity log type, and the height of the larger blocks is equal to the total probability of observing each activity log, summing across all temporal values. The actual inter-activity times represented by one symbol from each group are depicted below the graph.

Mentions: For each of the four activity log types (select, swap, reswap, and wiggle), all of the corresponding continuous-valued temporal features were clustered using K-means clustering. A set of centroids representing clusters of temporal values were identified, and the data points assigned to each cluster were re-coded with a corresponding cluster number. The K-means algorithm requires a priori specification of the number of clusters to generate from the dataset, and through experimentation we decided to use 15 clusters to represent the range of temporal features for each activity log. Each data sequence was re-coded using this vector quantization scheme, generating a new categorical observation vector composed of values from a 60-symbol alphabet, where each symbol represents both an activity log and temporal feature centroid. Figure 4 depicts a sample discrete observation probability distribution for the 60 possible observations. The height of each small bar corresponds to the probability of observing the symbol, and each bar corresponds to a centroid representing a cluster of inter-activity temporal feature values. The bars are sorted in order of increasing inter-activity time value (faster slower) from left to right for each activity type. The parameters describing each hidden state's observation probabilities characterize the behavior of the individual in that state in terms of what activities they performed and how quickly they performed them.


Modeling strategic use of human computer interfaces with novel hidden Markov models.

Mariano LJ, Poore JC, Krum DM, Schwartz JL, Coskren WD, Jones EM - Front Psychol (2015)

Example of a discrete observation probability distribution of the 60-symbol alphabet used to encode the activity logs and temporal features. Each small bar corresponds to a centroid representing a range of temporal inter-activity values. The height of the bar represents the relative probability of observing the corresponding symbol. The bars are grouped by activity log type, and the height of the larger blocks is equal to the total probability of observing each activity log, summing across all temporal values. The actual inter-activity times represented by one symbol from each group are depicted below the graph.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4490801&req=5

Figure 4: Example of a discrete observation probability distribution of the 60-symbol alphabet used to encode the activity logs and temporal features. Each small bar corresponds to a centroid representing a range of temporal inter-activity values. The height of the bar represents the relative probability of observing the corresponding symbol. The bars are grouped by activity log type, and the height of the larger blocks is equal to the total probability of observing each activity log, summing across all temporal values. The actual inter-activity times represented by one symbol from each group are depicted below the graph.
Mentions: For each of the four activity log types (select, swap, reswap, and wiggle), all of the corresponding continuous-valued temporal features were clustered using K-means clustering. A set of centroids representing clusters of temporal values were identified, and the data points assigned to each cluster were re-coded with a corresponding cluster number. The K-means algorithm requires a priori specification of the number of clusters to generate from the dataset, and through experimentation we decided to use 15 clusters to represent the range of temporal features for each activity log. Each data sequence was re-coded using this vector quantization scheme, generating a new categorical observation vector composed of values from a 60-symbol alphabet, where each symbol represents both an activity log and temporal feature centroid. Figure 4 depicts a sample discrete observation probability distribution for the 60 possible observations. The height of each small bar corresponds to the probability of observing the symbol, and each bar corresponds to a centroid representing a cluster of inter-activity temporal feature values. The bars are sorted in order of increasing inter-activity time value (faster slower) from left to right for each activity type. The parameters describing each hidden state's observation probabilities characterize the behavior of the individual in that state in terms of what activities they performed and how quickly they performed them.

Bottom Line: We further report the results of a preliminary study designed to establish the validity of our modeling approach.Pre- and post-task questionnaires probed for self-reported styles of problem solving, as well as task engagement, difficulty, and workload.Overall, we find that this novel approach to decomposing unstructured behavioral data within software environments provides a sensible means for understanding how users learn to integrate software functionality for strategic task pursuit.

View Article: PubMed Central - PubMed

Affiliation: The Charles Stark Draper Laboratory, Inc. Cambridge, MA, USA.

ABSTRACT
Immersive software tools are virtual environments designed to give their users an augmented view of real-world data and ways of manipulating that data. As virtual environments, every action users make while interacting with these tools can be carefully logged, as can the state of the software and the information it presents to the user, giving these actions context. This data provides a high-resolution lens through which dynamic cognitive and behavioral processes can be viewed. In this report, we describe new methods for the analysis and interpretation of such data, utilizing a novel implementation of the Beta Process Hidden Markov Model (BP-HMM) for analysis of software activity logs. We further report the results of a preliminary study designed to establish the validity of our modeling approach. A group of 20 participants were asked to play a simple computer game, instrumented to log every interaction with the interface. Participants had no previous experience with the game's functionality or rules, so the activity logs collected during their naïve interactions capture patterns of exploratory behavior and skill acquisition as they attempted to learn the rules of the game. Pre- and post-task questionnaires probed for self-reported styles of problem solving, as well as task engagement, difficulty, and workload. We jointly modeled the activity log sequences collected from all participants using the BP-HMM approach, identifying a global library of activity patterns representative of the collective behavior of all the participants. Analyses show systematic relationships between both pre- and post-task questionnaires, self-reported approaches to analytic problem solving, and metrics extracted from the BP-HMM decomposition. Overall, we find that this novel approach to decomposing unstructured behavioral data within software environments provides a sensible means for understanding how users learn to integrate software functionality for strategic task pursuit.

No MeSH data available.