The Pinocchio Theory

Two recent scientific articles help to illuminate the notion of decision, which for Whitehead is constitutive of all actual entities.

In the first place, Bjorn Brembs, who was one of the co-authors of a 2007 paper that suggested that fruit flies are able to generate spontaneous behavior that is not determined in advance either by genetic pre-programming or by environmental cues, has released a new paper in which he generalizes his argument. Brembs cites research by himself and others that points to the “common ability of most if not all brains is to choose among different behavioural options even in the absence of differences in the environment and perform genuinely novel acts.” That is to say, fruit files and other animals possess a sort of “free will.” Brembs dismisses, of course, what he calls “the metaphysical concept of free will,” i.e. the traditional Cartesian notion that is “inextricably linked with one variant or another of dualism.” But he also rejects strict determinism, both on account of quantum indeterminacy, and — more directly in biological terms — on the basis of the idea that, for animals, complete predictability of behavior is not viable. Any organism that reacted to stimuli in a completely predictable manner could all too easily be wiped out by predators who were able to anticipate these responses. Therefore, “predictability can never be an evolutionarily stable strategy. Instead, animals need to balance the effectiveness and efficiency of their behaviours with just enough variability to spare them from being predictable… Competitive success and evolutionary fitness of all ambulatory organisms depend critically on intact behavioural variability as an adaptive function. Behavioural variability is an adaptive trait and not ‘noise’.” All this suggests that motile animals, at the very least, have evolved mechanisms to generate behavioraly variability — action that is not pre-determined, and hence not predictable. Moreover, organisms are able to control the extent of this variability. In many circumstances, routine, habit, and “instinct” are the best strategies; but “faced with novel situations, humans and most animals spontaneously increase their behavioural variability.”

Brembs cites many examples of “self-initiated actions” (behaviors that are spontaneously and endogenously generated) in all sorts of animal organisms, and not just among vertebrates. He suggests that neural mechanisms have evolved which exhibit and exploit an “unstable nonlinearity.” These brain mechanisms are “exquisitely sensitive to small perturbations,” and they are irreducible to any binary alternative between “complete (or quantum) randomness and pure, Laplacian determinism.” This provides the basis for what Brembs calls a scientific concept of free will: one that is not an absolute, dualistic concept, but an immanent and relative one: “The question is not any more ‘do we have free will?’; the question is now: ‘how much free will do we have?’; ‘how much does this or that animal have?’. Free will becomes a quantitative trait.”

Brembs rightly draws philosophical conclusions from his argument, even though he disclaims being a philosopher. “Analogous to mutation and selection in evolution, the biological process underlying free will can be conceptualized as a creative, spontaneous, indeterministic process followed by an adequately determined process, selecting from the options generated by the first process. Freedom arises from the creative and indeterministic generation of alternative possibilities, which present themselves to the will for evaluation and selection. The will is adequately determined by our reasons, desires and motives—by our character—but it is not pre-determined.” From this point of view, free will requires something like a “self,” which is able to determine its own action; we may infer such self-willed action whenever “no sufficient causes for this activity to occur are coming from outside the organism.”

Free will does not, however, necessitate consciousness in the human sense. Fruit flies make decisions — they determine and generate their own behavior, to the limits that external constraints allow them to — without necessarily being “conscious” of making these decisions. Even among human beings, this is most likely the case. Brembs cites, in passing, Benjamin Libet’s experiments, which suggested, by means of testing neural responses, that human beings make decisions prior to being conscious of their decisions. Libet’s results have often been cited as disproving the existence of “free will”; but Brembs rightly says that, although these results discredit the “metaphysical” (dualist) notion of free will, they “are not relevant for the concept proposed here.” For what Libet showed was not that I do not make spontaneous or uncaused decisions, but rather that my “mind” makes these decisions, or my brain generates them, prior to my becoming consciously aware of them. Brembs’ empirically grounded notion of free will is entirely consonant with the argument — one metaphysically beyond the scope of Brembs’ paper, but which I would want to make on Whiteheadian grounds — that things like consciousness and responsibility are not the grounds or preconditions for decision or the exercise of free will, but rather the consequences (in some, but not necessarily all, cases) of making decisions and (thereby) exercising free will.

Brembs suggests that free will is an evolutionary adaptation of the nervous system; it would thereby be restricted to animal organisms. But what about biological entities that don’t have nervous systems (including plants, fungi, protists, and bacteria)? All these organisms have been shown to engage in various sorts of cognitive activities. “Plant cognition and behavior” has come to be a recognized biological subfield; bacterial “quorum sensing” is widely recognized and experimented upon; and slime molds (in particular, the model organism Physarum polycephalum) have been shown to exhibit “smart behavior” in solving a maze, and to solve “combinatorial optimization problems.” But most of this research has focused on cognition and problem-solving, not on the issue of free will that Brembs raises in connection with fruit flies and other invertebrates.

Slime molds are particularly interesting organisms, because they are neither unicellular nor multicellular, but something in between. They exist for most of their lives as blobs of protoplasm with many nuclei. Meiosis occurs at the end of the life cycle, when the slime mold develops “fruiting bodies” composed of haploid spores. These spores are widely dispersed, and begin their lives as haploid, single-celled organisms. Two of these unicellular organisms mate, forming a larger cell with a diploid nucleus. But from that point on, mitosis, or the separation and replication of nuclear DNA, is not accompanied by cell division. Rather the entire blob grows in size as it comes to contain multiple nuclei. The blob moves around, sending out filaments of protoplasm in various directions as it searches for food. It is in the course of this process, which seems not to be centrally coordinated, but to involve internal communication among different parts of the organism, that slime molds have succeeded in threading mazes and solving combinatorial problems. [I am referring here to myxomycetes, or “true” slime molds; as opposed to the also interesting, but vastly different, cellular slime molds].

[One might also note that Gilbert Simondon ponders at great length on the question of whether animals that live in colonies, like coral, are truely individuated or not. Is each organism an individual? Or is it only the colony that is an individual? Obviously, the same question could apply to the notion of ant or bee colonies as superorganisms. But the case of slime molds is even stranger; as far as I can recall, Simondon never mentions them (please, somebody, correct me if I am wrong). Slime molds are more than individual cells, but less than differentiated multicellular organisms. Not only don’t they divide into separate cells, but they don’t differentiate into separate tissues or organs, except when they form fruiting bodies at the point of sporulation. And, as mentioned above, this differentiation takes place, and the spores become separate entities, only via meiosis. This question is related to the fact, discussed below, that slime molds do not make decisions as unified “individuals,” but only as loose, decentralized collectivities — although, again, the members of this “collective” are not separate from one another, as they are in the cases of corals and of ants.]

This brings me to the second recent article I mentioned above. It concerns “irrational decision-making” processes in slime molds. This article, by Tanja Latty and Madeleine Beekman, is of much narrower scope than Brembs’ essay; and its explicit focus is entirely cognitive. Nevertheless, I think it is relevant to the questions that, following Brembs, I am raising. Latty and Beekman created situaitons in which slime molds were allowed to choose between different food sources, which varied both as to how nutritious they were, and as to how illuminated they were. Slime molds prefer richer food sources to poorer ones, but they also prefer darkness to light (since they are easily harmed by exposure to bright light and ultraviolet radiation). What “preferences” would the slime molds establish, when confronted by the alternative between a rich, but brightly-lit food source, and a poorer, but dimly-lit and therefore much safer one?

With multiple trials, and the insertion of additional alternatives, the scientists determined that slime molds, like human beings and other animals, do not operate in accordance with the dictates of what has been called (in the human social sciences) “rational choice” theory. That is to say, they do not make “economically rational” choices “based on the absolute value of items” they are choosing among, but rather “use comparative valuation rules.” There are many problems with rational choice theory, and even with the amended version, “behavioral economics,” which acknowledges that people (and other organisms that make decisions) often make use of “comparative valuation rules” and other, not-strictly-rational, cognitive shortcuts. I will not go into these problems here (that would require an entire separate essay, or several); suffice it to note that these approaches have an impoverished notion of “decision,” since they regard it not as spontaneously-generated activity, but merely as an ordered selection among items on a pre-determined menu or list.

Letty and Beekman don’t address Brembs’ question of free will, because they remain within an entirely cognitivist and behavioural-economic context. But two aspects of their experiments are nonetheless relevant here. In the first place, they suggest that the presence, among slime molds, of the same limited rationality and behavioral strategies that one finds among animals with nervous systems suggest that such strategies of choice are not just  “a consequence of the way brains process information,” but rather indicate “an intrinsic feature of biological decision-making,” even when brains and neurons are not involved. Although they (wrongly, in my opinion) regard decision in exclusively cognitive terms, as a form of information processing, they do not see this “processing” as an exclusively animal-based, or neurally-based activity, but give it a much wider provenance. We know that it is taking place in slime molds and other brainless organisms, even though we do not yet understand how this happens. This suggests that the biological basis of free will is not necessarily tied to neurons and nervous systems in the way that Brembs suggests; it is a broader, or more basic, evolved feature of organisms.

The researchers state that “acellular slime moulds, like insect colonies, are collective decision makers, where the behaviour of the collective is a result of the behaviour of its underlying parts. Each slime mould is made up of many tiny pieces of slime mould, each oscillating at a frequency determined partly by the local environment, and partly by interactions with adjacent oscillators such that each oscillator can entrain those close to it.”  Given this situation, and “owing to the slimy nature of acellular slime moulds, it was not possible to test [rationality] in individuals, and instead, we relied upon population-level preferences.” But there is still a weird difficulty here. The authors note that “recent work on rationality in ants,” in which each organism in a colony makes individual decisions, and the colony’s behavior as a whole is the sum of these decisions, “has led to the suggestion that organisms using collective decision-making processes may be immune to irrational decisions.” However, even if thisis the case with ants in a colony, it turns out not to be the case for slime molds. Is this perhaps because a slime mold is neither a unity, nor a collection of entirely separate individual units, but something strangely in between?

Another problem with rational choice theory and behavioral economic theory is that they assume separate individual “preferences” which are only summed secondarily and extrinsically. But in actuality,this is never the case. Every individual’s decisions are influenced by (even if not reducible to) the decisions of others, plus all sorts of supplemental contextual factors. As Whitehead says, in every process of decision “whatever is determinable is determined” by the situation in which the individual finds itself, the “stubborn fact” that it cannot evade; although at the same time “there is always a remainder for the decision” to be made by the actual entity itself (PR 27-28). This mixture of self-determination and dependence is a matter of degree, just like the balance between externally determined and internally self-generated action that Brembs describes. Slime molds represent an extreme ontological case, in which the contrast between internal and external definition, as well as between individual and collective determination, is pushed to its most intensely ambiguous point. This is why slime molds seem to slip in between the logic of separate individual decisions, and that of collective, but extrnisically-summed, decisions. Reducible to neither, they embody the point at which the logic of preferences-among-a-menu-of-items breaks down. And this is why Latty and Beekman’s focus on limited choice expands into something more like the indeterminacy of free will as defined by Brembs.

The second point I’d like to note from Latty and Beekman’s article is their finding that “even within a treatment group, slime moulds varied in their choices. This is particularly surprising as we controlled for weight, nutritional state and genetic differences.” In other words, even the slime molds’ compliance with “irrational” comparative valuation rules is not absolute. It is a statistical result, rather than something observed in every instance. This again suggests that there is a margin, or remainder, of indeterminacy that allows for unconstrained, spontaneous decision. The authors suggest that “some of the variability we observed arises from slight differences in the experiments’ initial conditions… These small differences in initial condition, combined with feedback via biomass recruitment mechanisms, could ultimately result in the observed variability.” This is undoubtably the case; but I would add that, as sensitivity to initial conditions approaches a point of indiscernibility, we get closer to Brembs’ claim that “determinism versus indeterminism is a false dichotomy,” which he bases in part on observing situations of extreme sensitivity to initial conditions. As Brembs puts it, “stochasticity is not a nuisance, or a side effect of our reality. Evolution has shaped our brains to implement ‘stochasticity’ in a controlled way, injecting variability ‘at will’.” The only amendment to this that we need to cover the case of slime molds is that this evolved ability to inject variability at will is not just a property of brains, but “an intrinsic feature” (as Letty and Beekman put it) of all biological entities.

I’ll end my own discussion here with a speculative epilogue that makes claims I cannot presently defend (although I am hopefully working towards them). It may be noted that research into biological free will and biological decision-making is not entirely unrelated to the questions about panpsychism raised by such analytic philosophers as Thomas Nagel, Galen Strawson, and Sam Coleman, and which I have discussed previously in this blog. For Strawson, the emergence of mentality from non-mentality is a serious problem, even though the emergence of life from non-life is not. He argues, therefore, that an incipient mentality must already exist on the level of subatomic particles. I suggest that it helps to make sense of this claim if we understand mentality in terms of “decision,” rather than in terms of consciousness or “qualia.” The evolution of biological decision making, and biological free will, might well depend upon, and make use of, an implicit potential of all matter. If decision were not already possible, then living things that actually do make decisions could not have come into existence. Rather than decision being a power of life, then, life would be a consequence of the potentiality of decision.