Related Experiment Video
Updated: Jul 17, 2025
07:05
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
Published on: October 27, 2016
9.3K
Self-orienting in human and machine learning
Julian De Freitas1, Ahmet Kaan Uğuralp2, Zeliha Oğuz-Uğuralp3
1Marketing Unit, Harvard Business School, Boston, MA, USA. jdefreitas@hbs.edu.
Nature Human Behaviour
|August 31, 2023
Summary
Humans excel at self-orientation, a key computational process for recognizing oneself in space and time. This ability, crucial for flexible navigation, surpasses current AI algorithms in simple self-finding tasks.
Area of Science:
- Cognitive Science
- Artificial Intelligence
- Computational Neuroscience
Background:
- A computational concept of self involves an agent recognizing its own body representation in a specific time and place.
- This self-representation is framed as a self-orientation process, posing a significant computational challenge for artificial agents.
- Self-orientation is fundamental for agents to effectively interact with and navigate their environment.
Purpose of the Study:
- To investigate the computational process of self-orientation in human agents.
- To compare human self-orientation capabilities with those of deep reinforcement learning algorithms.
- To explore the role of self-orientation in enabling flexible navigation.
Main Methods:
- Development of 'self-finding' tasks using simple video games.
- Recruitment of 124 human players to identify themselves within the game environment.
- Quantitative and qualitative assessments of player performance in self-orientation tasks.
- Evaluation of established deep reinforcement learning algorithms on the same tasks.
Main Results:
- Human players demonstrated near-optimal performance in self-orientation tasks.
- Deep reinforcement learning algorithms performed significantly below optimal levels.
- A notable discrepancy exists between human and AI capabilities in self-orientation, despite AI's success in complex games.
Conclusions:
- Humans possess a highly effective mechanism for self-orientation.
- Current deep reinforcement learning approaches are not optimized for self-orientation.
- Self-orientation is proposed as a critical component for enabling flexible environmental navigation in agents.
More Related Videos
Related Concept Videos
01:22Purposive Learning
140
E. C. Tolman emphasized the purposiveness of behavior — the idea that much of our behavior is goal-directed. For instance, employees who aim for a promotion work diligently to meet their targets. Tolman argued that when classical conditioning and operant conditioning occur, the organism acquires certain expectations. In classical conditioning, a child might fear a dog because they expect it to bite. In operant conditioning, a person might consistently work overtime because they expect a...
140
01:24Humanistic Psychology
1.1K
Humanistic psychology emerged in the mid-20th century as a response to the deterministic and pessimistic nature of behaviorism and psychoanalysis. While behaviorism focused on observable behaviors influenced by the environment and psychoanalysis delved into unconscious motivations, both theories suggested that human actions lacked free will. In contrast, humanistic psychology offers a perspective that emphasizes the innate potential for goodness and growth within every individual.
This approach...
This approach...
1.1K
01:20Introduction to Cognitive Psychology
525
Cognitive psychology is the field of psychology dedicated to examining how people think. It attempts to explain how and why we think the way we do by studying the interactions among human thinking, emotion, creativity, language, and problem-solving, as well as other cognitive processes. Cognitive psychology studies how information is processed and manipulated in remembering, thinking, and knowing.
This field emerged in the mid-20th century, following a period dominated by behaviorism, which...
This field emerged in the mid-20th century, following a period dominated by behaviorism, which...
525
01:18Introduction to Learning
470
Learning is the process of acquiring knowledge or skills through practice or experience, leading to long-lasting behavioral changes. This acquisition occurs through interaction with the environment and requires practice or experience. For instance, mastering a skill such as surfing requires considerable practice and experience, highlighting the essential role of repeated interactions with the environment in learning.
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...
470
02:16Stereotype Content Model
14.8K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
14.8K
01:27Associative Learning
439
Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...
Classical conditioning, also known...
439