Neuroforum 2018; 24(2): A95–A100 Martin Heisenberg Mind from Matter? – Via Brain and Behavior

Neuroforum 2018; 24(2): A95–A100 Martin Heisenberg Mind from Matter? – Via Brain and Behavior

Abstract: How did the process of Darwinian evolution lead from dead matter to the human mind? Of this long, complicated process the present essay selects and discusses just one step, that from animal behavior to animal mind. The process of living has two aspects, the maintenance of the process in the organism and the interaction of the organism with the world. In animals the latter is organized as behavior. Behavior evolves, as it serves the fitness of the animal. The brain evolves because it improves the behavior in terms of the animal’s fitness. Given the richness of the world and the openness of the future, the organization of behavior can be indirect and most intricate. The animal mind can be understood as behavioral organization at a higher level, as metaorganization. This concept is documented by behavioral studies in a particular animal, the fly Drosophila.

Keywords: Metaorganization; Darwinian evolution; biosphere; behavioral organization; animal mind

At the end of his scientific life Max Delbrück, one of the pioneers of molecular genetics, wrote a book (Delbrück, 1986) entitled: “Mind from Matter?” At that time most scientists considered the mind a part of human nature. The ‘Cartesian cut’ regarding mind and matter as two distinct ‘substances’ was out. Today the mind is thought to have appeared in this world as a process of Darwinian evolution.

As a molecular biologist one assumes that, to become alive, matter condensed and diversified from elementary particles to atoms, to molecules. Life must have started from aggregates of molecules that mutually catalyzed their synthesis. This required partitions to keep the relevant kinds of molecules together and to reduce variation. To become the first stable living organisms these aggregates of molecules still had to ‘invent’ precise duplication and division. Later came the metazoa with their cyclic regeneration from germ cells. Some of these creatures developed behavior, sociality, communication and one branch, the hominids eventually language and thought . This is about what Delbrück must have had in mind.

Delbrück’s book title ended with a question mark indicating how puzzled he still was regarding this scenario. He did not consider the question in the title still open. He wanted to remind us, that even if ‘Mind from Matter’ is true, it remains a fascinating question how the mind evolved. In the present article I take up Delbrück’s book title, focusing on one particular stage in this long evolutionary process, the role of brain and behavior. This, I hope, will make the arrival of the mind in evolution a bit more plausible.

By all we know, life is a unique process in the universe, confined so far to this planet. Life on earth, the biosphere, consists of myriads of organisms and their habitats. Most of them are single cells, each one encased by a membrane, a barrier separating it from the rest of the world. The process of life takes place to a large part in these tiny compartments. Every single one is alive, living by itself. Its organization is highly autonomous. The molecular processes in this aggregate keep the balance of life and the contact to the surround across the outer barrier.

Early organisms presumably did not have behavior. This probably evolved in metazoa, multicellular organisms. Perhaps behavior started with active mobility, which dramatically enriched the organism’s relations to the outside. Behavior evolved. It diversified, got more specialized, more complex. It needed to be organized. Behavior became the most significant aspect of an animal’s life.

The evolutionary perspective on behavior is so interesting, because it reminds us, that the outcome of behavior and thus also the behavior itself can be good and bad. Each creature is a separate experiment. In the evolutionary perspective the main quality of an organism is its fitness, its ability to transmit its genes to the next generations. For life to be good, the animal’s behavior has to be good. Behavior should contribute to the fitness of an animal. The organization of behavior is so demanding because its task is to optimize the outcome of behavior. To do the right thing, all life long.

Brains organize behavior. Brains evolved with the diversification of behavior and the deepening of its organization. The richer the behavior, the larger the brain. This relation is obvious. It is demonstrated for instance most drastically in the life cycle of certain tunicates. These animals have a mobile and a sessile form of life. In the mobile phase they make use of their brain (cerebral ganglion) but for the sessile phase they reduce it to a negligible rudiment. Also the fruit fly, which as a larva does little else but feeding, in this stage has a 10 times smaller brain than after metamorphosis, when its full behavioral repertoire has developed.

If we accept that brain is the kind of living matter that organizes behavior, we should return to a fundamental property of matter: its intrinsic initiating activity. This is of particular relevance in the brain. Brains are endogenously active. They can initiate a behavior spontaneously. Also behavior is mostly active. Behavioral organization can be modulatory. Animals can learn, try out. Brains can operate endogenously without being tied to an external stimulus. They can organize a potential future behavior independent of its actual context. They can adequately deal with the open future.

This gets me to the central topic of this essay: If the mind has come as a result of Darwinian evolution, it must have served the animal’s fitness. I would like to propose that the mind represents an indirect mode of behavioral organization, a level somewhat removed from the direct process of activating a behavior. With its mental abilities the animal prepares for potential future tasks of behavioral organization. Minds came into play, because the manifold of potential interactions between a mobile organism and the outside world is gigantic and because the comparative evaluation of behavioral options according to their future outcome is so involved.

To say it again, I consider mentality an indirect mode of behavioral organization. What do I mean by ‘indirect mode’? Imagine a publishing company. One might say the core business of the company is to print and sell books. Period. In reality, however, most of the activities of the company serve this task only indirectly. The lectors have to read the manuscripts and decide whether to publish them. The publisher invites the authors to discuss potential future book projects, he analyzes the various branches of the market, advertises forthcoming books, organizes reading tours for the authors, keeps contact with bookstores, libraries and consumers, designs a new logo for the company, acquires new production methods, etc, etc.. The core task, printing books and selling them is but a minor part of what is going on in the company.

In the same vein, the organization of behavior is served indirectly in many ways. Organisms have to interact with the outside world to survive and to generate offspring. A behavior is generated because of its effects. The effects will show in the future. The future is difficult to predict. The animal has to learn what the effects of its behaviors are under different circumstances. It has goals,

it tries out, it tries to understand perceptual phenomena, it learns about the space around it, other organisms, conspecifics, dangers, opportunities, in short: the world. Much of what brains do for behavior is mental. Thinking, feeling, learning, understanding improve behavior in one way or another, may-be not always, not right in the next round of action selection but more often than not.

For the remaining part of this essay I will describe examples of brain processes serving this indirect mode of behavioral organization. All the examples will be drawn from a small animal, the fly Drosophila. Mobile animals with several symmetrical pairs of appendages have a common ground plan of behavioral organization, even if they are as different as flies and primates. They need to get oriented in the world, have to find food, communicate, fight, copulate, solve problems, find shelter, hide, sleep, and many other activities. These commonalities in behavioral organization even show in brain structure, as the fossil record of a common ancestor (Fuxianhuia protensa) from 500 million years ago reveals (Strausfeld and Hirth, 2013). Whether all examples to be discussed must be classified as mental, is not the main issue. The examples are just to show that already in small brains like that of Drosophila behavioral control can be indirect, action selection can be highly involved. As brain and behavior of Drosophila are intensely studied, we know of many examples of higher-order control in this organism.

Motion vision: One of the best-studied fields of behavioral control in flies is visual flight control (Heisenberg and Wolf, 1984), in particular motion vision. The early work on motion vision, which had been conducted on larger flies (Reichardt and Wenking, 1969), had been paradigmatic for direct behavioral control. A visual stimulus reaches the eye and elicits a behavioral response. More precisely, the movement of the image of the surround on the fly’s eye elicits a syndirectional turning response that reduces the slip. This is the optomotor response. However, already this behavior is much more complex than just pushing a button for a bell. The brain has first to measure the direction of motion to activate the right behavior. Moreover, the control of optomotor behavior is much more complex than that. Only if the motion occurs outside in the world or is due to passive motion of the fly, the response is activated. If the retinal slip is caused by the fly’s own behavior, it does not elicit an optomotor response. In other words, the fly makes the action selection dependent also upon the cause of the stimulus rather than just the stimulus (Wolf and Heisenberg, 1986; von Holst and Mittelstaedt, 1950).

Orientation: Visual behavior is full of examples of higher-order control. The fly has a concept of the 360° azimuth space around it. For instance, it can remember a specific dangerous orientation and avoid getting close to it (Wolf and Heisenberg, 1997). A further example: A looming shadow over the fly’s head may elicit an escape jump. Before the take-off the fly assumes the right posture of its middle legs to direct the jump into the safest direction (Card and Dickinson, 2008).

Learning: As already shown in some of the examples above, action selection requires (pragmatic) knowledge of the world. Obtaining such knowledge involves learning and memory. This is one of the most intensely studied areas of brain and behavior research in Drosophila. Pragmatic knowledge may serve future behavior, if a corresponding situation arises. A hungry fly finds a food source emitting a certain odor and little later is exposed to a different odor without food. The next day the two odors are presented simultaneously from different directions. The fly turns towards the odor which the day before had been combined with the food. Why is this the indirect mode? Imagine the fly would escape the apparatus after the conditioning. It would most likely never get into a situation in which it would profit from what it had learned. In a natural habitat this particular odor selected for the experiment would hardly ever occur together with food (Chouhan et al., 2015).

Learning is costly. Therefore it is necessary to carefully choose which items of knowledge to store. Flies can remember the time of day something important happens for them, for instance, as just described, that in the afternoon but not in the morning a certain odor is associated with food. The flies establish time-of-day memory only if they are extremely hungry. Once this type of memory is stored, they go for the odor at the particular time of day, whether they are hungry or not. So, flies have a concept of the daily cycle and can label an event according to its phase in this cycle.

Female flies display a scent, which is irresistibly attractive for males. But if the male has to spend an afternoon together with a freshly inseminated female, which has no interest in males and this is likely to last for many days, he will lose interest in females for the days to come. So, there is another manifestation of the sense of time: time as a duration (Kelemann et al., 2007).

Intentionality: Humans often initiate a behavior because they have an intention. This is similar in flies. Flies have an expectation of the presumed outcome of the behavior they are going to activate and this expectation matters for the decision. One of the most intensely studied cases of intentionality is the approach towards a light source. Flies in a dark, narrow enclosure, if disturbed rush towards a light at the far end. This behavior is known as fast phototaxis. They do so, in order to escape into open space where they can fly away. How do we know? Because, if we temporarily immobilize their wings, they show no fast phototaxis. Flies need to continuously update their outcome expectations if they are going to use them in action selection (Gorostiza et al., 2015).

If a fly is motivated to climb across a cleft (Pick and Strauss, 2005), it first estimates the width of the cleft, next it strategically positions its hind- and middle legs, then it pushes its body forward over the cleft and stretches its front legs as far as possible to reach the other side. There the legs may extensively search for a hold. If the legs do not make it, the fly retreats. Occasionally it may even fall down into the cleft. If on the next day it happens to get to the same place, it may not even try again.

Intentionality in fly behavior rarely is as obvious as in behavioral sequences. However, one can readily study it experimentally. One can manipulate the outcome and show that the altered outcome is unexpected. The outcome must be represented in the brain regarding its value for the animal in comparison to other options and regarding its statistical validity. This may require to also learn about other aspects of the outcome (Heisenberg, 2015).

Learned helplessness: An example of outcome learning is an experiment called learned helplessness. The fly is enclosed in the so called shock-box, a narrow, completely dark alley, where it can walk or rest. Its position is recorded and it can receive mild but unpleasant electric shocks. If a shock arrives the fly tries to escape. If it is hit by the shock while resting, it immediately starts walking, while walking it turns around and walks the other way. This immediately turns off the shock. The fly shows this behavior without any prior experience. What, however, if all of a sudden the situation has changed and moving away does not terminate the shocks? If this happens again and again, the fly gives up the escape attempt. It learns that it can not do anything to control the situation and reduces the response frequency to a low level but not quite to zero. The loss of control has severe consequences. It causes a motivational state called ‘learned helplessness’ (Batsching et al., 2016). The fly walks more slowly, generates shorter walking bouts and longer pauses. It may also not perform in other learning experiments, as if it were depressed. Lithium and serotonin-related drugs are effective remedies (Ries et al., 2017). If the fly is undisturbed the frequency and duration of the walking bouts are controlled in a stochastic pattern by a brain center (central complex) that also organizes the fly’s orientation in space (Martin et al., 2001).

Attention: Flies show selective visual attention. In most instances a fly can perform only one behaviour at a time. It may happen, though, that different regions of the visual field may simultaneously display different matters of concern. In this case the fly will restrict its focus or window of attention to one of the regions and ignore the others. It can actively shift its ‘window’. Alternatively, the window can be cued to a different position by external stimuli, visual or non-visual (Sareen et al., 2011; Koenig et al., 2016).

Perceptual hypotheses: A sensory signal may have a meaning and a relevance for the fly. Often, however, stimuli may be ambiguous. A dark patch in the visual surround of a fly may be a predator or a place where to hide. If the first evaluation triggers ‘approach for hiding’, this decision may have to be revised if the patch starts moving.

Some ambiguities are stable in time. They are generally termed ambiguous figure rivalries. Well known examples in human vision are multi-stable percepts such as the Necker cube or Rubin’s vase. In these cases the subject looking at an image alternates stochastically between two equally plausible interpretations of what the image depicts.

Also Drosophila may experience sensory ambiguity and multi-stable perception (Toepfer et al., 2018). Confronted with a uniformly textured panorama in a virtual reality in stationary flight with the fly’s turning force coupled to the motion of the panorama (closed loop), the fly adjusts its yaw torque to stabilize its virtual self-rotation. To make the visual input ambiguous, one adds a second texture. Both textures get a rotatory bias, making them move in opposite directions at a constant relative angular velocity. The fly now has two possible frames of reference for self-stabilization: either of the two single textures. Results indicate that flies continuously alternate in their behavior adjusting their yaw torque to the mutually exclusive interpretations. The brain generates an interpretation of the stimulus and activates the appropriate behavior for this ‘hypothesis’. Yet, the outcome of the particular behavior is not really satisfying. So, again and again the animal tries the other interpretation.

Sleep: Flies sleep. They sleep mostly at night, but also occasionally during the day. They need sleep to consolidate long-term memory. If they are sleep-deprived for a night they make up for it with additional sleep the next day (Huber et al., 2004; Seugnet et al., 2008).

Motivational control: Emotional, dispositional states are typical higher-order elements of behavioral control (Su and Wang, 2014). They have already been mentioned several times. They are defined by their behavioral symptoms, in most cases with reference to possibly related states in humans (Anderson and Adolphs, 2014). Accordingly, flies can show states like pain, fear, zeal, tiredness, addiction (Kaun et al., 2012) and many others. They modulate the likelihood that a particular behavior is activated. Each one is involved in behavioral organization in many different ways.

Social interactions: Flies watch other flies and organize their behavior according to what their conspecifics do. Flies live longer, if they are not alone (Shohat-Ophir et al., 2012). The male uses wing vibration to generate a courtship song for the female. The female admits the male for copulation, if the song sounds right. Male flies fight with other males. If they loose, their chance to win the next fight is reduced (Ueda and Wu, 2009).

So much with examples. Most brain and behavior projects with animals currently under investigation relate to higher-order behavioral control. The examples chosen here are an arbitrary collection. As stated at the beginning, the fly’s behavior is the fly’s interaction with the world. Whether life is good or bad depends among others upon the effects of all these interactions. Knowledge of the world and tentative foresight of the effects of the behavior, goals, concepts of time and space, of causality, of self and non-self, of rivals, predators, shelter, of company with conspecifics, ect., ect., may all contribute to make behavior more successful.

Mind improves behavior. This is how mind could arise in evolution from matter. The question mark in Delbrück’s book title should now be a bit less puzzling, if one learns that mind arose from behavioral control and that the matter from which the mind evolved was the brain. If this indirect mode of behavioral organization would not have improved life and would not have supported the propagation of our genes, it would not have established itself in the living world. The examples of behavioral control in Drosophila showed how demanding it is to always do the right thing, throughout life. The mind makes behavior more successful. If in brain research we want to understand how the brain works, we must include studying the mind. Even with flies.

What made it so difficult for Delbrück to confirm, that mind arose from matter? At the end of the book he tells the reader why: “Did we explain how mind evolved from no mind? Did we find out why so much more was delivered than was ordered…? …does it even make sense to posit, that the capacity to know truth can arise from dead matter?” This should remind us, how little we still know about the mind and its evolutionary history. What were the large and small innovations? Obviously, the evolution of living matter was a fundamental prerequisite for the evolution of mind. How did behavioral organization evolve? How did the successive stages build upon each other? Did it start with active mobility or was oriented locomotion the primary demand? How important was social behavior and, in particular communication? To which extent is cultural evolution still Darwinian? Could even thinking be considered a higher level of behavioral control? ‘Mind from Matter’ still deserves many question marks.


Anderson, D. & Adolphs, A. (2014). A framework for studying emotions across species. Cell 157, 187–200.
Batsching, S., Wolf, R. & Heisenberg, M. (2016). Inescapable Stress Changes Walking Behavior in Flies – Learned Helplessness Revisited. PLoS ONE 11(11):e0167066. doi:1371/journal. pone.0167066.
Card, G. & Dickinson, M. H. (2008). Visually mediated motor
planning in the scape response of Drosophila. CurrBiol 18, 1300–1307.
Chouhan, N. S., Wolf, R., Helfrich-Förster, C. & Heisenberg, M. (2015). Flies Remember the Time of Day. CurrBiol 25, 1619–1624.
Delbrück, M. (1986). Mind from Matter? Blackwell Scientific Pub. Gorostiza, A., Colomb, J. & Brembs, B. (2015). A value-based behavioural choice underlies phototaxis in Drosophila. doi:
Heisenberg, M. (2015). Outcome learning, outcome expectations, and intentionality in Drosophila. LearnMem 22, 294–298. Heisenberg, M. & Wolf, R. (1984). Vision in Drosophila. Springer,Berlin.
Holst, E. von, & Mittelstaedt, H. (1950). Das Reafferenzprinzip. Wechselwirkungen zwischen Zentralnervensystem und Peripherie. Naturwissenschaften 37, 464–476.
Huber, R., Hill, S., Holladay, C., Biesiadecki, M., Tonioni, G.& Cirelli, C. (2004). Sleep Homeostasis in Drosophila melanogaster. Sleep 27, 628–639.
Kaun, K. A., Devineni, A. V. & Heberlein, U. (2012). Drosophila melanogaster as a model to study drug addiction. Hum Genet 131: 959–975.
Kelemann, K., Krüttner, S., Alenius, M. & Dickson, B. (2007).
Function of the Drosophila CPEB protein Orb2 in long-term courtship memory. NatNeurosci 10, 1587–1593.
Koenig, S., Wolf, R. & Heisenberg, M. (2016). Visual Attention in Flies – Dopamine in the Mushroom Bodies Mediates the After-Effect of Cueing. PLoS ONE 11(8): e0161412. doi.10.1371/ journal.pone.0161412.
Martin, J. R., Faure, P. & Ernst, R. (2001). The power-law distribution for walking-time intervals correlates with the ellipsoid-body in Drosophila. J. Neurogenet. 15, 205–219.
Pick, S. & Strauss, R. (2005). Goal-driven behavioral adaptations in gap-climbing Drosophila. CurrBiol 15, 1473–14878.
Reichardt, W. & Wenking, H. (1969). Optical detection and fixation of objects by fixed flying flies. Naturwissenschaften 56: 424–425.
Ries, A.-S., Hermanns, T. Poeck, B. & Strauss, R. (2017). Serotonin modulates a depression-like state in Drosophila responsive to lithium treatment. Nature Communications (8:15738) DOI: 10.1038/ncomms15738.
Sareen, P., Wolf, R. & Heisenberg, M, (2011). Attracting the Attention of a Fly. PNAS 108, 7230–7235.
Seugnet, L., Suzuki, Y., Vine, L.,Gottschalk, L. & Shaw, P. J. (2008). D1 receptor activation in the mushroom bodies rescues sleep-loss-induced learning impairments in Drosophila. Curr Biol 18, 1110–1117.
Shohat-Ophir, G., Kaun, K. R., Azanachi, R. & Heberlein, U. (2012).
Sexual deprivation increases ethanol intake in Drosophila. Science 335, 1351–1354.
Strausfeld, N. & Hirth, F. (2013). Deep homology of arthropod central complex and vertebrate basal ganglia. Science 340, 157–161.
Su, C.-Y. & Wang, J. W. (2014). Modulation of neural circuits: how stimulus context shapes innate behavior in Drosophila. Current Op in Neurobiol. 29, 9–16.
Toepfer, F., Wolf, R. & Heisenberg, M. (2018). Multi-stability with ambiguous visual stimuli in Drosophila orientation behavior. PLoS Biology 16(2): e2003113. pbio.2003113.
Ueda, A. & Wu, C.-F. (2009). Effects of social isolation on neuromuscular excitability and agressive behaviors in Drosophila: altered responses by Hk and gsts1, two mutations implicated in redox regulation. J Neurogenet. 23(4), 378–394.
Wolf, R. & Heisenberg, M. (1986). Visual orientation in Drosophila is an operant behavior. Nature 323, 154–156.
Wolf, R & Heisenberg, M. (1997). Visual space from visual motion: Turn integration in tethered flying Drosophila. Learning & Memory 4, 318–327.


Prof. Dr. Dr. h. c. Martin Heisenberg, Julius-Maximilians Universität Würzburg, Rudolf-Virchow-Zentrum, D15, Josef-Schneider-Straße 2, 97080 Würzburg, Germany, Phone: +49 931 3184451, Mail:

Martin Heisenberg, born 1940 in Munich, studied Chemistry, Biochemistry and Genetics in Munich and Tübingen. As a postdoc he worked with Max Delbrück, California Institute of Technology, Pasadena (1966–1968), before he joined the group of Karl G. Götz at the Max-Planck-Institute of Biological Cybernetics, Tübingen to study brain and behavior of the fly Drosophila melanogaster. He was one of the first to introduce Genetics as a tool in neuroethol-ogy. Since 1975 he is Professor at the University of Würzburg. His early studies of the fly visual system are summarized in a book “Vision in Drosophila” (1984; with R. Wolf). Other research foci are visual pattern recognition, the localization of memory traces, the role of initiating activity in operant behavior, perception, selective attention, and motivation. His goal is to establish a basic behavioral model of the brain.

Leave a comment

Your email address will not be published. Required fields are marked *

3 × = nine

Leave a Reply