I would like to put the Hierarchy out of the problem. In my view, the cooperation depends on motivation, level of expertise and task self-organisation of both sides. In my experience I have both: successful and unsuccessful cooperation with colleagues. If we deal professionally, I mean - without negative personal subjective attitudes or emotions (such as anger, selfishness, egoism), the work would be more effective. I am happy that I managed to have fantastic teamwork so far.

For the beginning, I want to present several typical and realistic cases of inefficiency of cooperation. Please comment on reasons and motivations.

1) Researcher A knows more than B, and B expects him to do most of the job.

2) Both of researchers A and B have high IQ and can write good papers. But they have different philosophy or methodology of research that cannot be used jointly in one paper. They argue a lot. In the end, it may take them less time to write papers separately than together.

So far I was talking about voluntary cooperation (endogenous). But there might be situations with exogenous cooperation. For example, an interdisciplinary business project is given to 3 institutions (A, B and C), known for their skills (without preliminary coordination between researchers). Suppose that A should create data used by B, then B should work out some algorithm and do calculations on the basis of those data and deliver it to C, while C should design some theory and work out policy implications.

Suppose that A (or B) fails to submit data (or algorithm for B) in time, and the overall deadline is approaching. This is a stressful situation. Suppose that A already received its fraction of research money, but B or C have to do their job in 2 weeks instead of 2 months, which is technically impossible. How to proceed?

People are generally used to hierarchies and to cooperate by them, for a monetary career reward. Although cooperating in networks is the foreseeable future, this communication method requires a different scientific mentality, e.g. open access, knowledge, motivation to respond, spontaneity, sharing, etc. We should also not forget the military and commercial implications, e.g .the silent commitment to sign certain 'documents'.

Dear , interesting issue.

I agree with Dr. Yuri Yegorov and Stephen I. Ternyik . To add some more:

Yes, the scientific collaboration has been implemented at national, regional and global lever for centuries order to broaden and deepen their knowledge and scope and exemplary achievements were recorded. Though there are astonishing achievements were recorded; however, sometimes inefficiencies in scientific cooperation are experienced at different levels. These are mainly due to :

• Miss use of science, violation of scientific integrity can be violated, and compormization of researchers. It is better to prevent this than to criticize and regret what has gone wrong.
• Researchers are not be able to work and cooperate freely and independently, without ulterior political or commercial motives. If that is not possible, it significantly affect the efficiencies of scientific collaboration through erosion of joining of forces.

If you look at the most creative periods in science, working together has invariably operated within compromise for the best results not within sustained, enforced hierarchies. Mid-European Physics from late 19th century until say 1960. Early 19th century until, say, 1870. Individuals appeared who changed perceptions. (I do not forget the 200 years of Arab or rather Moslem thinkers who's contribution was vast and operated in the same or similar fashion until forced to toe the line.) Hierarchies normally imply conservativism in thought.

But the above remarks are surely, or largely, concerned with laboratory research which is usually team work and in which fall-outs are common? The results of laboratory work are different, focused on different goals, with different activity therefore different co-operative methods are required.

I want to thank everybody for your answers which give complementary visions on the topic. But there are no sharings of personal experiences yet.

@Stephen points out the role of hierarchies. Indeed, they are common in most of research. But I also know many examples when people with similar status (both having PhD) write joint papers. They may be in the same or different institutions.

@Alexander reveals of important change: now quantity is valued higher than quality (because only future generations will see real quality). A good strategy is to split the task in several pieces and then to write several coauthored publications, instead of 1 paper for 1 researcher (like Goedel did in the past).

@Stanley says that I mean laboratory research. Not indeed. This research typically requires efforts of several people from one lab. In economics some purely theoretical papers can be the result of cooperation between different labs or even institutions.

Exchange programmes

I think a reason is "do not find stable points of accord".

Yuri, I guess I was contrasting independent effort and combined effort. In the past, scientists were more or less amateurs, while really that is not now the case. The greatest creativity has, surely undeniably, been in the previous setups, or within the concept of movable talent as in the second period I itemise, not perhaps in the present set-up.

Sorry, to say it, but I think there is more bad science now in the institutional set-ups.

I would like to put the Hierarchy out of the problem. In my view, the cooperation depends on motivation, level of expertise and task self-organisation of both sides. In my experience I have both: successful and unsuccessful cooperation with colleagues. If we deal professionally, I mean - without negative personal subjective attitudes or emotions (such as anger, selfishness, egoism), the work would be more effective. I am happy that I managed to have fantastic teamwork so far.

American sociologist of Russian origin P. Sorokin singled out two mandatory conditions for social interaction:

the participants in the interaction must possess the psyche and sense organs, ie, means that enable one to know what the other person feels, through his actions, facial expressions, gestures, voice intonations, etc .;

participants in the interaction should express their feelings and thoughts in the same way, that is, use the same symbols of self-expression.

Interaction can be considered both at the microlevel and at the macro level.

Interaction at the micro level is an interaction in everyday life, for example, within a family, a small work collective, a student group, a group of friends, etc.

Interaction at the macro level unfolds within the framework of social structures, institutions and even society as a whole.

Depending on how the contact is made between interacting people or groups, four main types of social interaction are distinguished:

physical;

verbal, or verbal;

non-verbal (mimicry, gestures);

mental, which is expressed only in inner speech.

The first three refer to external actions, the fourth to internal actions. All of them have the following properties: meaningfulness, motivation, focus on another person.

***

- In the psychological sense: positive interaction, in which the goals and interests of the participants coincide or the achievement of the goals of some participants is possible only through ensuring the interests and aspirations of its other participants.

***

Robert Cecile: "It seems that the governments of many nations want the suicide of the universe, while the peoples are hungry for peace" (from the speech of September 3, 1936).

https://en.wikipedia.org/wiki/Robert_Cecil,_1st_Viscount_Cecil_of_Chelwood

Bertrand Russell: "A person can not be at peace with others until he learns to be at peace with himself ... The future horrified me, but even more horror I felt because 90 percent of the population happily anticipates a bloodbath ... If only people work so hard to make themselves happy, as they are now working to make others unhappy, the world would be a wonderful place "(XX century).

https://en.wikipedia.org/wiki/Bertrand_Russell

Niels Bohr: "Only on the basis of harmonious cooperation between peoples the fate of future generations must be built ... The person is now engaged in such problems that his spirit captures and makes him dizzy. However, until you feel a slight giddiness, you will not be able to understand their essence. Problems are more important than solutions. Decisions can become obsolete, and problems remain ... Scientists usually argue to reach an agreement or at least agree on certain issues. The purpose of political discussions seems to be to achieve the highest degree of disagreement "(XX century).

https://en.wikipedia.org/wiki/Niels_Bohr

I want to thank everybody for new points in our discussion.

@Alexander, I agree that theoretical sciences and math are more individual than experimental (like most of physics, chemistry, biology). Articles in math that have more than 1 author are mostly a product of hierarchy. But I know also about cooperation between scientists of the similar level. Very often one knows one part better, and other another part. It can be a cooperation between algebra and geometry, functional analysis and theory of differential equations, etc. Such cooperation develops interdisciplinary links inside mathematics. In the last decades it is often the case when one person writes formulae, and the other does numerical simulations, but it is often a hierarchy, because older generation in stronger in theory, while younger in computer skills.

@Ekaterina, I agree that cooperation depends on motivation from both sides, and I assume it to exist. Still there might be some failures, based on psychological interaction (as @Zabrodskii mentions) or disagreement on some basic principles (as @Mariano suggests).

But what about cheating? To what extent it can destroy a potentially fruitful cooperation, in other case mutually beneficial? One form of cheating can be non-revealing the level of one side knowledge resulting in its free-ride on the capacities on the other side.

I assume than in honest cooperation both sides deliver the best of their knowledge, and avoid the possibility to bring negative effect to their partner. I also mean by "cheating" the case when it does not hold.

Such a situation can be modeled as a game in economics. Suppose that one side has more power (financial or other) and other more knowledge. Then there is a possibility of manipulation using this power. Power can also arise from game structure, like the side doing the first step under time constraint has more power, and extract it from strategic delay. In practice, one side can manipulate the timing of delivery of its input that is crucial for the work of the 2nd side. Clearly, such "cooperation" cannot result in long and fruitful cooperation, but as a one-shot game can give advantages to the powerful side (at the cost of joint efficiency).

I can also recommend to visit my other question about the relative role of power (will) and knowledge: https://www.researchgate.net/post/What_is_more_important_for_success-knowledge_or_will?

Yes, Alexander, I do not know a paper in math with 10 coauthors. But I am a participant of a paper in applied math with 6 co-authors. See Article Placing the poor while keeping the rich in their place

Indeed, such discussions and cooperative work were mutually beneficial. There was also good psychology. In applied math you do not only solve some problem, but you also formulate it. And there are many alternative set ups. Some of them lead to more realistic description, and other to mathematically interesting case. Discussions results in an optimal set up. Then you have to go analytically as far as it is possible, and after that use a computer. Simulation in multi-parameter space is also tricky.

This is an important question with not so obvious answers. In addition to the excellent observations already made, there is a bit more to add.

The distinction between vertical cooperation (a Mentor and his students) and horizontal cooperation (cooperation between colleagues) is very helpful. In both cases, the success of a cooperation depends upon the extent that those participants who have insights that strengthen (add to) the solution of a research problem. Here are examples of both types of cooperation that have been successful:

Vertical cooperation. Georges Matheron and Ph.D. student Jean Serra, 1964 launched the successful Mathematical Morphology (MM), which made it possible to identify the holes and objects in the foreground of a digital image. Another remarkable achievement of MM is the mapping of a digital image containing various objects to the contours of the objects. With MM, it is also possible to carry out a deformation retraction from an object contour to a skeleton. Sample object contours and retractions of object contours to skeletons are shown in the attached images. For more about this, see

http://cmm.mines-paristech.fr/~serra/pdf/birth_of_mm.pdf

Horizontal cooperation. Socrates and his students Plato and Xenophon provide a remarkable example of a successful horizontal cooperation. Socrates was born circa 470 BC, in Athens, Greece. We know of his life through the writings of his students, including Plato and Xenophon. His "Socratic method," laid the groundwork for Western systems of logic and philosophy. For more about this, see

https://www.biography.com/people/socrates-9488126

Dear @James, thank you for examples of real cooperation. If image recognition principles have been known back in 1964, I imagine how far better they are today. Facebook manages to recognize our faces even from low quality photos.

Dear @Alexander, you provide almost a typical example about bosses projecting power by obliging people to sign their names in papers without real contribution. Perhaps typical for more corrupted societies, but I am not sure if all Western universities are completely free from that.

Dear @Peter, yes, power matters a lot. But sorry, I did not get your idea about cooperation with your colleague, if you and he always do the opposite. Is this no cooperation example? What are the benefits for both?

P.S. By the way, following RG suggestion, I added a pre-print today. It is written about quite general issues (also about valuation of rare pieces of art, real estate and wine), and I would be glad to see interesting comments. See Preprint Valuation of Unique Objects: The Role of Heterogeneity and Information

In my opinion, failurs haben when the two researches not close (different) in thinkink and ideas, espically in new cooperation.

I know the case when the famous physicist Lev Lanadau, the Nobel Prize winner in physics in 1962, who has a disgusting character, perfectly cooperated with his colleague Evgeny Livshits and even wrote together many volumes on theoretical physics. Why? Because Landau did not know how to write by accessible and understandable language, however, Livshits did it wonderfully. Landau have been discussing another topic with Livshits, and then Livshits wrote the discussion on paper until Landau disagreed with the final variant, after which everything was repeated for another topic. However, it is unlikely that Landau would have done this if he could write like Livshits. He would write one without co-authors. Thus, the main reason for successful cooperation in my understanding is "one complements the other and together is a single whole". The most common examples are:

1. "Your money + My brains = Monography"

2. "Your Entrepreneurship + My Science = Textbook"

3. "Your laboratory + My knowledge = Discovery" and so on.

Nevertheless, a failure is possible with a high probability under the formula "My brains + Your brains = Failure", that is, in a situation where two ambitious scientists are going to do joint work.

For very effective scientific cooperation or collaboration there must be willingness, desire and interest to work on common topic or related subject matter. The researchers must be ready to contribute almost equally to the problem at hand. Sacrifice is also an essential factor.

@ Aleksandar Jovanovic:

i did not see your examples earlier!? On line editors in RG?

Why did I not know those things on MM work...

The collaboration between Matheron and Serra is remarkable. It started in 1964 and continued until roughly 2000 (36 years). Online MM course materials are available at

http://cmm.ensmp.fr/~serra/cours/index.htm

For the MM research center, see

Centre de Morphologie Math\'{e}matique, EMP, Fontainebleu:

http://en.wikipedia.org/wiki/Paris_School_of_Mines

More to the point, Matheron and Serra illustrate a hugely beneficial collaboration. Ironically, Jean Serra was a civil engineer with a B.Sc. in Gestalt psychology. By contrast, Georges Matheron (see his attached picture) was a French mathematician and geologist:

https://en.wikipedia.org/wiki/Georges_Matheron

Dear @Gennady, you have a good example. All physicists know Landau-Livshitz books. This example shows that complementarity is important. Even if it is asymmetric. Landau had unique skills in this pair, and perhaps could find many coauthors like Livshitz. But he needed somebody, lacking an important skill to write clearly for students.

Your other examples are correct, but they are not from science. It is a pity that scientific talents often lack money, laboratory, etc - and need to trade co-authorship for what they lack.

I know many examples of cooperation between 2 brains, and indeed was asking for examples of failures. Yes, lack of common point in mutual work, intention to free ride, lack of complementary skills, bad character(s) - all these factors can be reasons for failures.

@ Aleksandar Jovanovic: membership - if a pix belongs to the contour or is outside in dynamic conditions...

For raster images, the contour of an object is a simple, closed curve containing pixels on the boundary of the object shape. A pixel not on the boundary is in the interior of an object shape or is not part of the particular shape. Non-boundary shape pixels are ignored by MM in finding a shape contour.

From a topological perspective and after triangulating an image, it will probably be helpful to consider Alexandroff nerve complexes, where the nucleus of the nerve contains a shape boundary pixel. In that case, for a given nerve nucleus, the nerve would overlap a shape contour.

Good question

When working with members of a team from the same discipline there is more empathy and better results, but when working with a multidisciplinary group, it is first necessary to establish the functions and work of each member so that the expected result is achieved.

I agree with my teachers to answer

Dear Yuri, the "cheating-power" situation is in fact happening all the time, isn't it? Here we have the need to make/or not to make compromises. For example, a leading scientist, project coordinator who makes serious efforts to run the project successfully, participating in several joint publications with small contribution. When we pass through institutional evaluations, we usually calculate the percentage of our individual contribution. Sometimes it is fair, sometimes - not. In personal relationship and cooperation it is hard to achieve such precision, but if we could behave professionally - it is possible, and could be instrumentally useful. The personal choice is to make clear distinction in the very beginning, to "negotiate" afterwards, or to divide the contribution equally.

I agree with Dr. Ekaterina Luybenova Markova

I think that efficient and inefficient or successful and unsuccessful scientific cooperation depends on the unity and clear vision of goals in addition to existing of motivation.

Dear Ekaterina, you set up correctly possible cheating. It might be an unjust system, but it is not the break of cooperation. If the power is asymmetric, but everybody wants to continue cooperation, it is not the break up of it, although it might be less efficient. But here we have some administrative framework behind cooperation.

I agree with Mahmoud that motivation and common goals are key factors for a successful cooperation. Especially in the case of free cooperation, when there are no institutions that force scientists to cooperate (and I was talking mostly about such cases).

In some cases there are both positive and negative effects. It is much easier to write a paper with a co-author, who easily understands and complements you. Here even the pieces of work can be distributed. But if 2 co-authors know the subject equally well but push the article's objective (or style) towards different directions, a lot extra efforts are spend on coordination, a lot of written text is deleted, and the final outcome may be not mutually satisfactory.

But in some cases co-authors may open doors to different directions of model elaboration, and the final outcome will have higher quality (despite higher efforts).