Minds of small things · Article

Trial and error

Trial and error is the idea that an organism meets a difficulty by producing varied movements and keeping those that relieve it. H. S. Jennings gave it this form for simple animals in 1906.¹ Driesch, Loeb and Child questioned or contrasted it with other accounts.³⁴⁶ Later authors use the term for learning and for evolution.⁷⁸

Earliest held: 1906, Jennings, H. S.
Most discussed in: Behavior of the Lower Organisms, 1906
In the library: 179 passages in 15 works
Rewritten: 2026-10-06

Jennings's formulation

In Behavior of the Lower Organisms (1906), Jennings named three significant features of behavior. One was reaction by varied or overproduced movements, with selection from the conditions those movements produce.¹ Jennings called this reaction by selection of overproduced movements and described it as the constructive one of two mainsprings for the development of behavior.¹ Its effectiveness depends in part on how fit the movements are to relieve the organism from unfavorable conditions.² Jennings gave an example: against a diffusing chemical, the rapid movements of Paramecium are more likely to save the animal than the slow motion of Amoeba.²

Objections from Driesch and Loeb

Driesch (1908) found the term appropriate for an experiment described by Yerkes, where a series of trials and errors becomes an immediate reaction on a second stimulus.³ Driesch thought it unjustified where there is no objectified experience, since no criterion shows that a state is liked in lower animals.³ Loeb rejected the idea more sharply. Of the claim that worms move by heliotropism in strong light and by trial and error in weak light, Loeb compared it to saying a stone falls by gravity while a feather finds its way down by trial and error.⁴ Loeb also reported that Jennings and Mast put all such orientations under trial and error, and asked whether Euglena was closer to random selection or to the direct orientation of amphipods.⁵

Later uses of the term

Child (1924) contrasted the two patterns. Trial and error depends on the absence of mechanisms for direct orientation, while the tropism theory postulates such mechanisms in all reacting organisms.⁶ Watson and colleagues (2016) describe reinforcement learning as modifying connections at random and keeping changes that improve the output, and say this is equivalent in effect to natural selection.⁷ Newman (2019) argues that inherency of form and function does much of the work the standard model gives to trial-and-error-based natural selection.⁸ Trewavas (2025) cites Jennings and later studies for the claim that Stentor uses trial and error learning.⁹

SourcesEach quotation was checked word for word against the passage it opens.

reaction by varied or overproduced movements, with selection from the varied conditions resulting from these movementsJennings, H. S., 1906 · Behavior of the Lower Organisms · open at passage 872
the relative fitness of the movements to relieve the organism from the unfavorable conditionsJennings, H. S., 1906 · Behavior of the Lower Organisms · open at passage 854
unjustified to speak of trial and error when there is no objectified experienceDriesch, H., 1908 · The Science and Philosophy of the Organism · open at passage 243
as erroneous as the assertion that while a stone falls under the influence of gravity a feather finds its way down by the method ofLoeb, J., 1918 · Forced Movements, Tropisms, and Animal Conduct · open at passage 278
include all such orientations under trial and error and contrast them with the direct orientation of such animals as the amphipodsLoeb, J., 1918 · Forced Movements, Tropisms, and Animal Conduct · open at passage 118
the trial-and-error method of reaction to any particular factor depends on the absence of mechanisms which provide for immediate, direct, or in the higher animals and man, conscious and intelligent orientationChild, C. M., 1924 · Physiological Foundations of Behavior · open at passage 595
This can be as simple as modifying connections at random and retaining modifications that improve the output.Watson RA, Mills R, Buckley CL, Kouvaris K, Jackson A…, 2016 · Evolutionary Connectionism: Algorithmic Principles Underlying the Evolution of… · open at passage 36
Inherency clearly does much of the work attributed in the standard model to trial-and-error-based natural selection.Newman SA, 2019 · Inherency of Form and Function in Animal Development and Evolution · open at passage 70
Stentor uses trial and error learning in constructing intentional behaviourTrewavas A, 2025 · Plant intelligence dux: a comprehensive rebuttal of Kingsland and Taiz · open at passage 6

Linked ideas

Behaviour of single cellsrelated to

Jennings builds the idea from the movements of Paramecium and Amoeba, and compares how effective their differing movements are at escaping unfavourable conditions.

Tropismin tension with

Child's note pairs the tropism theory (Loeb) with the trial-and-error theory (Jennings) as rival accounts of animal behaviour.

Criteria of mindrelated to

Washburn's most common criterion, modified behaviour after individual experience, is the kind of evidence trial-and-error learning supplies.

Habit and plasticityrelated to

Washburn's maze studies concern how habits form and how earlier habits help or hinder new ones, the learning side of habit.

Determinism in physiologyin tension with

Loeb says life phenomena unequivocally determined by physicochemical conditions yield constant quantitative results, and rejects animal movements being voluntary or dictated by Jennings' 'trial and error'; Child describes it as random movements.

Decision-making and learning in slime mouldrelated to

The Y-maze rewards food at one branch and measures later choices, a learning test comparable to trial-and-error work.

Memory in planariaprecursor of / follows

Jennings read the planarian's head-waving and testing movements as varied behaviour that samples conditions, an early account of behaviour that can later be modified.

Vitalism and mechanismrelated to

Loeb and Driesch each treated the term as a test of whether behavior needs more than physicochemical explanation.

Physical forces and formparallels

Thompson uses trial and error as a drawing method for fitting coordinate systems to sections, a procedural sense of the phrase.

Conditioned reflexparallels

Operant and instrumental learning depend on consequences, while classical conditioning pairs stimuli.

Exploratory adaptation and random-network searchparallels

Exploratory adaptation is a search by random variation and retention, much as trial and error is, though here the search runs in connection strengths.

Natural selection and the struggle for existencerelated to

Selection among varied candidates and trial-and-error within an organism are both processes of keeping what works.

Forced movements and compensatory orientationin tension with

Forced-movement accounts leave little room for trial and error, and Jennings's stress on direction of motion points the other way.

Free energy principle and active inferencerelated to

Fields and Levin note that minimizing free energy requires probing the environment by perturbing it, which resembles exploratory trial behaviour.

Explore-exploit trade-off and foraging heuristicsrelated to

Sampling options and then committing to one is a structured form of trial and error.

Inherency of form, mesoscopic materials and the physical origin of body plansin tension with

Newman argues that inherency, not gradual trial-and-error selection, accounts for much of the evolution of form and function.

Basal cognitionrelated to

Trewavas uses Jennings's trial and error in protozoa as evidence for intelligence in simple organisms.

Computational design automation and in-silico search of morphologyin tension with

Blackiston and colleagues contrast random trial-and-error search of design space with following gradients toward optimal designs.

Affordance and action selectionin tension with

Reinforcement learning is described as learning by trial and error and reward, which the active inference accounts claim not to need.

Where it is discussedPassages matching trial and error, avoiding reaction, selection of movements, overproduced movements

1906	Behavior of the Lower Organisms · Jennings, H. S.	133
1924	Physiological Foundations of Behavior · Child, C. M.	13
1908	The Animal Mind: A Textbook of Comparative Psychology · Washburn, M. F.	9
1918	Forced Movements, Tropisms, and Animal Conduct · Loeb, J.	5
1916	The Organism as a Whole, from a Physicochemical Viewpoint · Loeb, J.	3
1908	The Science and Philosophy of the Organism · Driesch, H.	3
2016	Evolutionary Connectionism: Algorithmic Principles Underlying the Evolution of… · Watson RA, Mills R, Buckley CL…	2
2025	Plant intelligence dux: a comprehensive rebuttal of Kingsland and Taiz · Trewavas A	2
2019	Inherency of Form and Function in Animal Development and Evolution · Newman SA	2
1912	The Mechanistic Conception of Life · Loeb, J.	2