Ontolog Forum
| Session | Cognition |
|---|---|
| Duration | 1 hour |
| Date/Time | 03 Jun 2026 16:00 GMT |
| 9:00am PDT/12:00pm EDT | |
| 5:00pm BST/6:00pm CST | |
| Convener | Ken Baclawski |
Ontology Summit 2026 Cognition
- Arun K. Majumdar and John F. Sowa Reasoning Beats Pattern Matching
- For over 60 years, the best AI reasoning was based on the four step cognitive cycle: abduction, deduction, evaluation, induction, and repeat. Abduction generates hypotheses or educated guesses. Deduction derives implications. Evaluation chooses the best option. Induction combines the result with previous knowledge.
Many versions of the cognitive cycle have been invented and named. For guiding fighter pilots, John Boyd called it the OODA loop: Observe, Orient, Decide, Act. He originally said that each step would be traversed in milliseconds, but he later applied the loop to design and analysis steps that may take minutes, hours, or days. Whatever the time scale, the four steps are fundamental to reasoning in science, business, and life.
The pattern matching methods of Large Language Models (LLMs) are superb for translating languages, natural or artificial. They are also good for finding and relating patterns in large volumes of data of any kind. That enables them to answer questions by finding information or by applying previous methods to new data. For many problems, pattern matching can discover abductions or educated guesses. But deduction and evaluation cannot be done unless a similar cognitive cycle can be found somewhere on the WWW.
With the VivoMind system from 2000 to 2010, the authors used conceptual graphs for symbolic reasoning about a wide range of problems. For the new Permion system, they added LLM pattern matching to map conceptual graphs to and from natural language. But pattern matching, by itself, cannot do any reasoning unless it can find and adapt an appropriate cycle on the WWW. It often requires a huge amount of searching even for relatively simple examples.
In summary, LLMs cannot do reasoning unless and until the system finds a suitable cognitive cycle on the WWW. But the Permion reasoning methods automatically do the four-step cycle. If necessary, they can also do LLM searching, but none is required. - Video Recording
- YouTube Video
- For over 60 years, the best AI reasoning was based on the four step cognitive cycle: abduction, deduction, evaluation, induction, and repeat. Abduction generates hypotheses or educated guesses. Deduction derives implications. Evaluation chooses the best option. Induction combines the result with previous knowledge.
Conference Call Information
- Date: Wednesday, 03 June 2026
- Start Time: 9:00am PDT / 12:00pm EDT / 6:00pm CEST / 5:00pm BST / 1600 UTC
- ref: World Clock
- Expected Call Duration: 1.5 hour
- Video Conference URL: https://us02web.zoom.us/j/86994661673?pwd=mMUeaWyWhBMSzTw3SgH5GjMv2Qx4rH.1
- Meeting ID: 869 9466 1673
- Passcode: 803090
- Please download and import the following iCalendar (.ics) files to your calendar system.
Discussion
Josh Lieberman 12:30 The Burns quote is no longer a joke since present chatbots are dangerously good at faking sincerity and empathy.
Wesley Spacebar 12:32 except that they can't make eye contact yet
Josh Lieberman 12:35 (Edited) You jest, but a huge number of people are using chatbots as defacto therapists because they are convinced they’re being “seen”.
Wesley Spacebar 12:36 well, LLMs don't have the best intersubjectivity in relationships, but they are trained to behave nicely by smart people in SF
Alican Tuezuen 12:38 Arun, are you using some sort of standard for the prompt or how did you come up with that "structure" of the prompt?
Nanjangud Narendra 12:46 Are Conceptual Graphs used here? If so, how?
Hussein Ezzeldin (FDA/CDER/OMP) 12:49 what is the difference between agents?
Gary Berg-Cross 12:53 What foundational models do you use?
Marcia Zeng 12:54 Can you demo your answer with any knowledge graph (KG)?
Gary Berg-Cross 12:56 Is common logic used to store the developed ontologies?
Simon Polovina 12:57 Can you demo the use of CGs?
Marcia Zeng 13:00 FYI: There is a demo in a previous session recorded: https://youtu.be/6K6F_zsQ264
Gary Berg-Cross 13:02 Is abductive reasoning used in the examples you showed? Do you have an abductive agent? So you could ask the system show this reasoning workflow
Gary Berg-Cross 13:11 At this point We can probably agree that your system is not an agent in the world and doesn't have real feelings but the idea is that you might expand on a BDI belief desire intention model and create a model that connects some likely feelings to an agent that includes BDI aspects. Is there a plan to connect to and use an app like Mathematica for math computation?
Resources
Previous Meetings
| Session | |
|---|---|
| ConferenceCall 2026 05 20 | Education |
| ConferenceCall 2026 05 13 | Interoperability |
| ConferenceCall 2026 05 06 | Interoperability |
| ... further results | |
Next Meetings
| Session | |
|---|---|
| ConferenceCall 2026 06 10 | Education |
| ConferenceCall 2026 06 17 | Synthesis |
| ConferenceCall 2026 06 24 | Symposium |