Learning and cognition in highspeed decision making
It is widely accepted that more time and information yield better decisions. However, some decisions manage to be extremely fast and yet accurate. The trick of such highspeed decisions appears to be the use of simplifying heuristics that works well for the most common condition but lacks flexibility...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
eLife Sciences Publications Ltd
2025-06-01
|
| Series: | eLife |
| Subjects: | |
| Online Access: | https://elifesciences.org/articles/99634 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850157982512316416 |
|---|---|
| author | Martin Krause Wolfram Schulze Stefan Schuster |
| author_facet | Martin Krause Wolfram Schulze Stefan Schuster |
| author_sort | Martin Krause |
| collection | DOAJ |
| description | It is widely accepted that more time and information yield better decisions. However, some decisions manage to be extremely fast and yet accurate. The trick of such highspeed decisions appears to be the use of simplifying heuristics that works well for the most common condition but lacks flexibility otherwise. Here, we describe an unexpected level of flexibility in a complex highspeed decision that is made faster than an Olympic sprinter can respond to the start gun. In this decision, archerfish observe the initial speed, direction, and height of falling prey and then use these initial values to turn right towards where ballistically falling prey would later land. To analyze the limits in flexibility of this highspeed decision, we developed and critically tested a system that allowed us to replace the usual ballistic relation between initial prey motion and the expected landing point with another deterministic rule. We discovered that, surprisingly, adult fish could reprogram their highspeed decision to the new rule. Moreover, after reprogramming their decision fish were immediately able to generalize their decision to novel untrained settings, showing a remarkable degree of abstraction in how the decision circuit represented the novel rule. The decision circuit is even capable of simultaneously using two distinct sets of rules, one for each of two visually distinct objects. The flexibility and level of cognition are unexpected for a decision that lacks a speed-accuracy tradeoff and is made in less than 100 ms. Our findings demonstrate the enormous potential highspeed decision making can have and strongly suggest that we presently underappreciate this form of decision making. |
| format | Article |
| id | doaj-art-833a796a3ff241ff8d3b5d8c092bae4f |
| institution | OA Journals |
| issn | 2050-084X |
| language | English |
| publishDate | 2025-06-01 |
| publisher | eLife Sciences Publications Ltd |
| record_format | Article |
| series | eLife |
| spelling | doaj-art-833a796a3ff241ff8d3b5d8c092bae4f2025-08-20T02:24:00ZengeLife Sciences Publications LtdeLife2050-084X2025-06-011310.7554/eLife.99634Learning and cognition in highspeed decision makingMartin Krause0https://orcid.org/0009-0008-8884-9522Wolfram Schulze1Stefan Schuster2https://orcid.org/0000-0002-0873-8996Department of Animal Physiology, University of Bayreuth, Bayreuth, GermanyDepartment of Animal Physiology, University of Bayreuth, Bayreuth, GermanyDepartment of Animal Physiology, University of Bayreuth, Bayreuth, GermanyIt is widely accepted that more time and information yield better decisions. However, some decisions manage to be extremely fast and yet accurate. The trick of such highspeed decisions appears to be the use of simplifying heuristics that works well for the most common condition but lacks flexibility otherwise. Here, we describe an unexpected level of flexibility in a complex highspeed decision that is made faster than an Olympic sprinter can respond to the start gun. In this decision, archerfish observe the initial speed, direction, and height of falling prey and then use these initial values to turn right towards where ballistically falling prey would later land. To analyze the limits in flexibility of this highspeed decision, we developed and critically tested a system that allowed us to replace the usual ballistic relation between initial prey motion and the expected landing point with another deterministic rule. We discovered that, surprisingly, adult fish could reprogram their highspeed decision to the new rule. Moreover, after reprogramming their decision fish were immediately able to generalize their decision to novel untrained settings, showing a remarkable degree of abstraction in how the decision circuit represented the novel rule. The decision circuit is even capable of simultaneously using two distinct sets of rules, one for each of two visually distinct objects. The flexibility and level of cognition are unexpected for a decision that lacks a speed-accuracy tradeoff and is made in less than 100 ms. Our findings demonstrate the enormous potential highspeed decision making can have and strongly suggest that we presently underappreciate this form of decision making.https://elifesciences.org/articles/99634 blink decisionurgent decision makingrule-based decision makingspeed-accuracy tradeoffheuristicsreflex |
| spellingShingle | Martin Krause Wolfram Schulze Stefan Schuster Learning and cognition in highspeed decision making eLife blink decision urgent decision making rule-based decision making speed-accuracy tradeoff heuristics reflex |
| title | Learning and cognition in highspeed decision making |
| title_full | Learning and cognition in highspeed decision making |
| title_fullStr | Learning and cognition in highspeed decision making |
| title_full_unstemmed | Learning and cognition in highspeed decision making |
| title_short | Learning and cognition in highspeed decision making |
| title_sort | learning and cognition in highspeed decision making |
| topic | blink decision urgent decision making rule-based decision making speed-accuracy tradeoff heuristics reflex |
| url | https://elifesciences.org/articles/99634 |
| work_keys_str_mv | AT martinkrause learningandcognitioninhighspeeddecisionmaking AT wolframschulze learningandcognitioninhighspeeddecisionmaking AT stefanschuster learningandcognitioninhighspeeddecisionmaking |