Transient versus asymptotic memory in bistable systems
Memory in a dynamical system can be transient or asymptotic. Transient memory is held for a time and then forgotten. Asymptotic memory persists because the system has more than one stable state, so where it ends up depends on where it started1. A bistable switch is the plain case, recording its last switching event2. The distinction lets authors compare how much of its past a cell retains3.
- Earliest held
- 2008, Wolf DM, Fontaine-Bodin L…
- Most discussed in
- Memory in microbes: quantifying…, 2008
- In the library
- 27 passages in 8 works
- Rewritten
- 2026-10-12
Two kinds of memory
Wolf and colleagues (2008) described history-dependent behavior as spanning several time scales. Short-term, or transient, memory is stored for some time and then forgotten. A system may have effective long-term memory when its transient dynamics are long compared with environmental fluctuations. It has true asymptotic memory when its stationary state depends on initial conditions, as in nonlinear systems with multiple attractors1. Their example of the latter is the bistable switch, whose state encodes the last switching event2.
Measured in bacteria
Wolf and colleagues applied the distinction to B. subtilis. Different cellular systems remember different aspects of prior history, so the pathways together store more about the past than any one pathway. The total asymptotic memory was still somewhat less than the total transient memory, at 1.57 against 1.96 bits3. Fernando and colleagues (2009) analysed a circuit model of associative learning in single cells and found a bistable switch with an off and an on state. Its conditioned response lasted about 10.5 hours, then was lost as w2 fell below the threshold4.
Bistable switches in cells
Lyon (2015) notes that memory storage in animals with nervous systems was long thought to rely on a kinase and phosphatase pair acting as a bistable switch5. Law and Levin (2015) modelled resting potential bistability and concluded that bistable switches may be built by overexpressing particular channels in oocytes and mammalian cells6. Levin (2019) argues that voltage-gated channels readily form loops that amplify small signals or give robustness, without needing gene-regulatory mechanisms7.
Attractors and history
Power and colleagues (2015) take up the same vocabulary for ecosystems. In reply to commentary they agreed that multiple attractors are a generic property of dynamical systems. What they claim as interesting is the location of the attractors with respect to the history of the system8. They describe a system that converts alternate ecosystem states into alternate community states, turning patterns from past environmental states into ecological memories9.
‘true’ asymptotic memory if the stationary state of the system depends on initial conditions, as occurs in nonlinear systems with multiple attractors
Wolf DM, Fontaine-Bodin L, Bischofs I, Price G, Keasling J…, 2008 · Memory in microbes: quantifying history-dependent behavior in a bacterium · open at passage 12the state of a bistable switch encodes an asymptotic memory of the last switching event.
Wolf DM, Fontaine-Bodin L, Bischofs I, Price G, Keasling J…, 2008 · Memory in microbes: quantifying history-dependent behavior in a bacterium · open at passage 12the total asymptotic memory is still somewhat less than the total transient memory (1.57 vs. 1.96 bits).
Wolf DM, Fontaine-Bodin L, Bischofs I, Price G, Keasling J…, 2008 · Memory in microbes: quantifying history-dependent behavior in a bacterium · open at passage 40the conditioned response continues for approximately 10.5 hours, by which time [w2]=22.8 nM. Any longer and it is lost, as the level of w2 goes below the threshold.
Fernando CT, Liekens AM, Bingle LE, Beck C, Lenser T…, 2009 · Molecular circuits for associative learning in single-celled organisms · open at passage 17Memory storage in animals with nervous systems was long thought to involve a mechanism involving an autophosphorylating protein kinase and paired phosphatase that operate together as a bistable switch
Lyon P, 2015 · The cognitive cell: bacterial behavior reconsidered · open at passage 76Bistable switches may be constructed by overexpressing Kir2.1 in amphibian oocytes and Nav1.6 in mammalian cells, and by limiting expression of many of the non-inward-rectifying forms of voltage-gated potassium channels.
Law R, Levin M, 2015 · Bioelectric memory: modeling resting potential bistability in amphibian embryos… · open at passage 41it is not the attractors per se that we claim are interesting – it is their location with respect to the history of the system.
Power DA, Watson RA, Szathmáry E, Mills R, Powers ST…, 2015 · What can ecosystems learn? Expanding evolutionary ecology with learning theory · open at passage 58Figure 6 shows a system that converts alternate “ecosystem states” into alternate “community states”; thus converting patterns from past environmental states into ecological memories.
Power DA, Watson RA, Szathmáry E, Mills R, Powers ST…, 2015 · What can ecosystems learn? Expanding evolutionary ecology with learning theory · open at passage 43Such channels readily enable loops which either amplify small signals (positive feedback) or implement robustness (negative feedback)
Levin M, 2019 · The Computational Boundary of a "Self": Developmental Bioelectricity Drives… · open at passage 70
| 2008 | Memory in microbes: quantifying history-dependent behavior in a bacterium · Wolf DM, Fontaine-Bodin L, Bischofs I… | 17 |
| 2015 | What can ecosystems learn? Expanding evolutionary ecology with learning theory · Power DA, Watson RA, Szathmáry E, Mills… | 4 |
| 2016 | Evolutionary Connectionism: Algorithmic Principles Underlying the Evolution of… · Watson RA, Mills R, Buckley CL… | 1 |
| 2015 | The cognitive cell: bacterial behavior reconsidered · Lyon P | 1 |
| 2019 | The Computational Boundary of a "Self": Developmental Bioelectricity Drives… · Levin M | 1 |
| 2015 | Bioelectric memory: modeling resting potential bistability in amphibian embryos… · Law R, Levin M | 1 |
| 2022 | Probabilistic Inference with Polymerizing Biochemical Circuits · Katz Y, Fontana W | 1 |
| 2009 | Molecular circuits for associative learning in single-celled organisms · Fernando CT, Liekens AM, Bingle LE… | 1 |