Cognitive Bias Detection in Self-Reported Data Using Predictive Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to effectively detect and correct cognitive biases in self-reported data, which can lead to inaccurate analysis and interpretation of physiological and psychological data.
Innovation Solution
A system and method for detecting and correcting cognitive biases in self-reported data by generating an ontology of bias descriptor features, using predictive models to evaluate and adjust data, and recommending changes to device, application, or study design to mitigate biases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If self-reported data is collected from users, then data quantity and ease of collection are improved, but cognitive biases are introduced that reduce data reliability
Solution Approach 1:
The patent introduces an intermediary system comprising predictive models and bias descriptor features that mediate between the self-reported data and the final analysis. This intermediary layer detects and corrects cognitive biases without eliminating the self-reporting mechanism, thus preserving data collection efficiency while improving reliability.
Solution Approach 2:
The system implements feedback by using predictive models to generate expected values and comparing them with actual self-reported data. The bias descriptor features provide feedback signals that identify deviations caused by cognitive biases, enabling corrective adjustments to the data.
2Reliability
If predictive models are used to detect cognitive biases, then data reliability is improved, but system complexity increases
Solution Approach 1:
The patent changes parameters by introducing bias descriptor features that characterize different types of cognitive biases. These features transform the analysis from raw data comparison to a structured evaluation based on predefined bias characteristics, making the complexity more manageable and systematic.
Solution Approach 2:
The system segments the bias detection process into distinct components: predictive models generate expectations, bias descriptor features identify specific bias types, and correction mechanisms address individual bias instances. This segmentation reduces overall system complexity by breaking down the complex task into manageable modules.
Data Source
AI summary
Embodiments are provided for cognitive bias detection and correction in self-reported data. In some embodiments, a system can include a processor that executes computer-executable components stored in memory. The computer-executable components include first components that creates an ontology of bias descriptor features to identify cognitive biases. The cognitive biases can include a combination of at least one device-induced cognitive bias, at least one testing-application-induced cognitive bias, or at least one study-design-induced cognitive bias.