Adaptive HMI Control for Cognitive Training Adherence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cognitive training methods for modifying perceived value and behavioral responses to physical objects, such as the Go/NoGo task and Cued Approach Training, are limited by short training sessions, lack of individual adjustments, and non-engaging task environments, and do not fully control for user expectations, leading to minimized efficacy and adherence issues.
Innovation Solution
A human machine interface (HMI) system that controls cognitive training by repeatedly rendering physical objects associated with predefined criteria, monitoring user inputs, and updating performance metrics to enhance motivation and adherence, while personalizing tasks based on user preferences and progressively adapting difficulty levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional cognitive training tasks (Go/NoGo or CAT) are used with short training sessions and limited individual adjustments, then the task simplicity and ease of implementation are maintained, but the training efficacy and user adherence are minimized
Solution Approach 1:
The training task is divided into two distinct components: a Go/NoGo task for response inhibition training and a Cued Approach Task for attentional bias modification. Each component can be independently configured and adjusted, allowing complex training objectives to be achieved through modular task segments rather than a single complex task
Solution Approach 2:
The system dynamically adapts task parameters based on user performance and preferences. Difficulty levels, stimulus types, and task parameters are automatically adjusted during training sessions, transforming a static simple task into a dynamic adaptive system that maintains optimal challenge without requiring complex manual configuration
2Ease of operation
If conventional cognitive training uses non-engaging task environments, then the implementation simplicity is maintained, but the user adherence and motivation are reduced
Solution Approach 1:
The system allows flexible modification of task parameters including stimulus types (food images, geometric shapes, etc.), reward structures, time limits, and difficulty levels. These parameter changes enable the creation of engaging customized training environments without fundamentally altering the core task structure, maintaining implementation simplicity while enhancing user adherence
Solution Approach 2:
The system implements real-time feedback mechanisms through performance monitoring, score tracking, and adaptive difficulty adjustment. Users receive immediate feedback on their performance, and the system automatically adjusts subsequent task parameters based on this feedback, creating an engaging interactive experience that improves adherence while relying on automated processes rather than complex manual operations
3Device complexity
If conventional behavioral decision training does not control for user expectations, then the task design simplicity is maintained, but the causal inferences on training effectiveness are confounded
Solution Approach 1:
The system implements expectation control specifically in the Cued Approach Task component, where cues are selectively presented for target items versus control items. This localized application of expectation control in a specific task component allows for rigorous causal inference without requiring complete redesign of the entire training system, maintaining overall task design simplicity while improving measurement precision
Data Source
Figure 1~2
Figure 3
Figure 4A
AI summary
A technique for controlling a human machine interface, HMI, for a first task and for a second task is provided. The first task comprises outputting (404) a predefined criterion applicable to each of a plurality of physical objects. Each object is associated with a category within a group of at least pairwise disjoint categories with the criterion fulfilled for each object in a first category and not fulfilled for each object in a second category. Controlling (420) the HMI for the first task comprises repeatedly performing the steps of rendering an object; monitoring the HMI for an input during a predefined first time period after the rendering of the object; and updating a first metric indicative of a performance measurement in the first task. The method further comprises rendering (406) a plurality of receptacles each enclosing one of the objects for a second task. Controlling (430) the HMI for the second task comprises repeatedly performing the steps of rendering an object at any one of the receptacles for a predefined second time period; selectively rendering a cue at the rendered object within the second time period; monitoring the HMI for an input responsive to the rendering of the cue within the second time period; and updating a second metric indicative of a performance measurement in the second task.