Overview
Analytical Connectionism is a 2-week summer course on analytical tools, including methods from statistical physics and probability theory, for probing neural network models of higher-level cognition. The course brings together neuroscience, psychology and machine-learning communities, and introduces attendees to analytical methods for neural network analysis and connectionist theories of higher-level cognition and psychology.
Connectionism, a theoretical approach in cognitive science, uses neural network models to simulate a wide range of phenomena, including perception, memory, decision-making, language, and cognitive control. However, most connectionist models remain, to a certain extent, black boxes, and we lack a mathematical understanding of their behaviours. Recent progress in machine learning theory has provided novel analytical tools that have advanced our mathematical understanding of deep neural networks, and have the potential to help make these “black boxes” more transparent.
During the School, teams of students work closely to develop research projects on topics related to analytical connectionism, presenting initial proposals during week one and interim results at the School’s conclusion.
Additionally, students are grouped based on their expertise and preferences and assigned to take notes for a specific lecturer. These notes are peer-reviewed and collected into a special issue, creating a permanent resource for the community and giving note-takers the opportunity to contribute to a formal publication.
This course will introduce:
- mathematical methods for neural network analysis, providing a solid overview of the analytical tools available to understand neural network models;
- key connectionist models and empirical findings from developmental psycholinguistics, which provide targets for analytical results.
During the course, you will:
- attend lectures given by leading researchers on theoretical methods and applications, key connectionist models, and experimental observations;
- participate in tutorials, poster sessions, and contributed talks;
- present to and engage with lecturers, organisers, and other participants during a poster session;
- collaborate in a group with other participants on a novel research project, mentored by the course organisers and lecturers;
- work together and co-author lecture notes for publication in a special issue.
Language acquisition
How can a learner infer linguistic knowledge from sensory data? How is syntactic, lexical, phonological, and other linguistic knowledge integrated over the course of development? These questions are seeing a resurgence of interest due to the capabilities of large language models.
This year’s school will bring together theoretical scientists, computational and experimental neuroscientists, and psycholinguists, and introduce analytical tools for understanding this process in high-dimensional, non-linear systems.
Topics will include:
- analytical frameworks for understanding language development from statistical physics;
- empirical findings from developmental psycholinguistics with existing computational frameworks;
- emergent theories to identify statistical properties of linguistic environments that enable learning;
- understanding the emergence of syntactic and semantic representations in high-dimensional spaces.
Dates
| Applications open | 13 Feb 2026 |
| Application deadline | 8 May 2026 |
| Outcome communicated | 26 May 2026 |
| Deadline to accept admission | 14 Jun 2026 |
| School begins | 17 Aug 2026 |
Schedule
| Time | Mon 8/17 | Tue 8/18 | Wed 8/19 | Thu 8/20 | Fri 8/21 |
|---|---|---|---|---|---|
| 08:30 | Registration | ||||
| 08:45 | Initiative seminar intro | School intro | |||
| 09:00–10:30 | NNs learn to encode structure with TPRs Paul Smolensky | TPRs enable processing of discrete structure Paul Smolensky | NECST theory: Gradient symbolic computation Paul Smolensky | Generative AI & generative linguistics Paul Smolensky | Unsupervised and supervised prototype-based learning Michael Biehl |
|
10:30–11:00
break
|
|||||
| 11:00–12:30 | The puzzle of language acquisition Caroline Rowland | NECST theory: Compositionality and grammar Paul Smolensky | Example: Learning of a linearly separable rule Michael Biehl | The emergence of grammatical representations Caroline Rowland | Recent developments, extensions and challenges Michael Biehl |
|
12:30–14:00
lunch
|
|||||
| 14:00–15:30 | The statistical physics of learning Michael Biehl | The structure of children's input Caroline Rowland | Layered neural networks for regression Michael Biehl | The many paths to language Caroline Rowland | Child language acquisition in the age of LLMs Caroline Rowland |
|
15:30–16:00
break
|
|||||
| 16:00–17:30 | Poster session | Spotlight talks | Project brainstorms | Project brainstorms | Project organisation |
| 18:00– | Speaker dinner | School dinner at Restaurang Tvåkanten | |||
| Time | Mon 8/24 | Tue 8/25 | Wed 8/26 | Thu 8/27 | Fri 8/28 |
|---|---|---|---|---|---|
| 09:00–10:30 | Curriculum learning in natural and artificial intelligence Chris Summerfield | From correlations to grammar Francesco Cagnetta | Looking for a semantic core in LLM representations Marco Baroni | Abstract generalization in continuous representation models Isabel Papadimitriou | Project work |
|
10:30–11:00
break
|
|||||
| 11:00–12:30 | Word-object learning via visual exploration in space Larissa Samuelson | Lessons from high-dimensional asymptotics of two-layer networks Bruno Loureiro | How learning and use shape evolving linguistic systems Kenny Smith | Teaching a next-token predictor to reason Luca Saglietti | Project work |
|
12:30–14:00
lunch
|
|||||
| 14:00–15:30 | The role of input in shaping linguistic generalizations Kanishka Misra | The lifelike language learner Alex Cristia | Play-Doh's problem Asad Sayeed | Modeling human cognition in the age of LLMs Afra Alishahi | Project presentations |
|
15:30–16:00
break
|
|||||
| 16:00–17:30 | Project work | Project work | Project work | Project work | Project presentations |
| 18:00– | Speaker dinner | Speaker dinner | |||
- Day 1 (17 Aug) takes place at Studion, A Working Lab, Campus Johanneberg. From Day 2 onward (18-28 Aug), lectures take place in room 2123 Energin at Natrium.
- School dinner at Tvåkanten, about 15 minutes away from the lecture hall.
Lecturers
Core lecturers
Frontiers lecturers
Participants
Organisers
Directors
Practicalities
Target audience
This course is appropriate for graduate students, postdoctoral fellows, and early-career faculty in psychology, neuroscience, physics, computer science, and mathematics. The course is limited to no more than 40 participants to ensure a balance across fields. Attendees are expected to have a strong background in one of these disciplines and to have made some effort to introduce themselves to a complementary discipline.
We strongly encourage students from underrepresented groups to apply and will offer bursaries to applicants who face financial barriers. In circumstances where all other things are equal, priority will be given to qualified applicants from underrepresented groups.
Course fees
There is a course fee of 2,500 SEK including VAT (2,000 SEK excluding VAT). Attendees are expected to cover their own travel, accommodation and other subsistence expenses. Lunch and coffee/tea breaks will be provided on course days, and there will be one course dinner.
Cancellation policy:
- Within 7 days of registering/making the booking: 100% refund
- After 7 days of registering/making the booking and up to 15 days before the start of the school: 80% refund
- Less than 15 days before the start of the school: no refund
Financial assistance via travel grant may be available for successful applicants who find it difficult to take up a place for financial reasons. Applicants are asked to indicate in their application if they would like to be considered for financial aid. The amount of financial aid available will depend on the course funding from grants and sponsors.
Frequently asked questions
- Will I receive individual feedback if my application is not successful?
Unfortunately, due to the large number of applicants, we are unable to offer individual feedback on unsuccessful applications.
- I need a visa to attend the programme. Will you provide support?
Sweden is part of the Schengen Area. If you require a Schengen visa, we will provide an official invitation letter to support your application after you have accepted your place. Please begin the visa application process as soon as possible, as processing times can be long.
- Where can I stay in Gothenburg?
Gothenburg offers a range of accommodation options. In addition to listings on platforms such as Airbnb and Booking.com, we have arranged discounted rates for participants at the following locations:
- Holtermanns - double rooms 800 SEK/night (can be shared by two); family rooms 1,200 SEK/night (can be shared by three).
- Göteborgs Vandrarhem (hostel) - 6-8-bed rooms (gender-separated) 395 SEK/person/night; 4-bed rooms 475 SEK; 2-bed rooms 550 SEK.
- Comfort Hotel Panorama - single at 980 SEK/night VAT incl. Accepted participants can book at a discounted price following this link valid until June 16th.
- How do I get from Gothenburg Landvetter Airport to the city?
Vy Flygbussarna runs the direct airport coach to central Gothenburg, with several city stops before terminating at Gothenburg Central Station (Nils Ericson Terminalen). The full journey takes approximately 35 minutes. Departure frequency varies, so please check the current timetable close to your arrival.
Your best onward connection will depend on where you are staying and on traffic conditions. Use the Västtrafik journey planner or the Västtrafik To Go app to plan the rest of your journey. Please note that the airport coach is a separate service and is not included in a regular Västtrafik ticket.
- How should I travel around Gothenburg, and which ticket should I buy?
Gothenburg has an extensive network of trams, buses, trains and ferries. We recommend downloading Västtrafik To Go for route planning, live service information and tickets: Android (Google Play) or iPhone (App Store).
We recommend the 30-day Zone A season ticket, which is sufficient for both School venues and allows travel in Gothenburg and the surrounding area, including a large part of the archipelago. Standard 30-day tickets are temporarily discounted during 2026. Check the Västtrafik To Go app
- Do I need cash in Sweden?
Sweden is largely cashless, and debit and credit cards are accepted almost everywhere. A physical card or a card in your phone’s wallet is generally the easiest way to pay.
If you have not bought a Västtrafik ticket in the app, you can tap a contactless Visa or Mastercard on the reader aboard buses, trams and ferries in Zone A to buy one adult single ticket. Each traveller needs a separate payment card. For journeys outside Zone A, use the Västtrafik To Go app to buy the correct ticket.
- I have dietary requirements. Will catering accommodate them?
Yes, we ask for dietary requirements as part of registration, and we will pass them on to the caterers.
- What are the requirements for poster presentations?
Due to space constraints, posters must be in A1 portrait format (594 mm × 841 mm).
- What are the requirements for spotlight presentations?
Presentations will be short, maximum five minutes. We will email final guidelines to all presenters closer to the programme start date.
- Why is the course fee now listed as 2,500 SEK?
The course fee is 2,000 SEK excluding VAT. We have now been informed that VAT must be added, making the total amount payable 2,500 SEK including VAT. The previously listed amount did not include VAT, and we apologise for the confusion.
- What is included in the course fee?
The fee includes lunch and coffee/tea breaks on course days, as well as one course dinner. Attendees are expected to cover their own travel, accommodation and other subsistence expenses.