Psychological experiments are less 'law-like' than physics or biology to the degree that the variables involved are harder, if not in principle impossible, to control. Inversely, physical or chemical 'laws' are only true if very specific ceteris paribus conditions ("all else being equal") hold. The former is not a mark of 'failure' or 'impotency' on the ledger of psychology so much as it is an acknowledgement of the specificity of the subject matter under investigation. A law-like psychology would not be a psychology at all - or simply bad psychology.

So there's nothing very special about 'laws'- in fact they are rather peculiar, holding only in very controlled, very un-natural situations. Laws are in fact about as close to idealizations as you can get (following the philosopher of science Nancy Cartwright, the more interesting question, frankly, is whether or not the law of physics are themselves true. Cartwight's position, which is more or less mine, runs: "rendered as descriptions of facts, they are false; amended to be true, they lose their fundamental, explanatory force"). That an account of meaning ought to be codified according to some law-like structure is, as such, plainly ridiculous. Frankly, I'd deem it beneath refutation had you not tagged me here.

Frankly, I'd deem it beneath refutation had you not tagged me here. — StreetlightX

Honestly, does everything you don't agree with have to be followed with a snarky remark about being "pedestrian" "stupid" and "beneath you"? I would be less inclined to want to spew vitriol if you didn't sour the waters so quickly and for no good reason other than being smug and promoting yourself as superior in all philosophical thought.. Has the efficacy of internet philosophy been reduced to how smarmy you can be to your interlocutor in a quick one-liner? (Please don't be ironic and try to have a one-liner answer in response :-} )

Anyways..That being said, I don't see how you properly solved what I see to be a dilemma with trying to use psychological conclusions for a basis in verifying philosophical theories. You seem to simply be saying that psychology will never verify be verifiable in the same (what you call "un-natural") situations as physics and biology. That, it seems we agree on.

I don't see then how this conclusion we agree on doesn't then lead to the idea that psychology's conclusions are mirky at best, and thus its veracity questionable as in justifying philosophical conclusions. Instead of agreeing with this conclusion, you went the other way and attacked physical "laws" because they are so peculiar and non-mirky (excepting things like Quantum Mechanics which in my opinion is a different sort of improbability than the improbabilities of psychology which are rather about the nature of contingency of events rather than a fundamental aspect of how particles act).

The ability to reproduce results consistently based on reported finding is one of the main problems with psychology. The research psychologist is under pressure to come up with new findings, not get embroiled in issues having to do with reproducible of results (perhaps a cultural issue?).

An August 2015 University of Virginia study looked at 100 psychological experiments published in three major psychological journals. The study attempted to replicate the results(contacting the original experimenters to insure their replication was fair): 35 of the studies held up, 62 did not and the researchers excluded 3 studies. The Virginia study noted that none of their attempts to replicate, contradicted the results of the experiments but that their results were significantly weaker than the original study had indicated. The study also looked at the 'prestige' of the research team and it did not matter. The only thing they found that mattered was the robustness of the results, experiments with very clear results were typically replicate able. http://www.nytimes.com/2015/08/28/science/many-social-science-findings-not-as-strong-as-claimed-study-says.html?_r=0

Kinda dismal results, especially since working psychologists depend on research for insight into their patients treatment.

The point is that psychology is 'murky' precisely in accordance to the degree in which it's subject matter is murky. If were any more precise, it would do injustice to what it studies. One doesn't measure the genetic robustness of a fish by it's ability to climb a tree, and neither should psychology be deemed somehow inadequate for being unable to codify it's findings in a law-like fashion. That said, what this has to do with philosophy is beyond me. You seem to want to say that because psychology is 'murky' (false), that this somehow places a burden on philosophy. But this already presupposes that philosophy ought to be undergridded by some sort of verificationist methodology which is nowhere spelled out, elaborated, or argued for in your OP - just assumed. And without that argument, I simply have no dog in the fight you seem to want to pick.

My snarkiness is simply a function of my disinterest in the OP, Schop. I'm not sure I can put "I don't care" in a delicate manner.

This is exactly what I am talking about. Thank you for the link and the information on the study. I agree with you that this is a problem with psychology in general but also with using psychology to back-up philosophical conclusions. It makes realist accounts of concept/object metaphysics especially hard to verify being that scientific realism depends on a large part on evidence captured at this level- evidence that is not reliable.

My snarkiness is simply a function of my disinterest in the OP, Schop. I'm not sure I can put "I don't care" in a delicate manner. — StreetlightX

I rather have that than what you presented earlier. I will not be offended.. Maybe someone's interest will be piqued. I just thought of you because the idea came when reading the Brassier article and our discussion in that thread. No problems if you aren't interested.

But this already presupposes that philosophy ought to be undergridded by some sort of verificationist methodology which is nowhere spelled out, elaborated, or argued for in your OP - just assumed. — StreetlightX

Not that you will respond, as you aren't interested, but I personally don't think that a verificationist methodology should be undergirding philosophy. However, many forms of realism are tied to science so heavily that it presupposes that the veracity of the science it is tied to be reliable. Being that realism has a much of its efforts tied in with science and experimentation, this would be something to consider.

did anyone test the results of the test, to see if the test results hold up? Perhaps the testers didn't understand the methodology of half the studies and were unable to reproduce results because of bad reproduction of methodology.

For instance, in one study this is included:

Differences from the Original

In addition to the original study design, we added the following question: “Are you a vegetarian or a vegan?” We planned to perform a one-way ANOVA to see whether there is a difference in the total number of thoughts reported by vegetarians/vegans compared to non-vegetarians/non-vegans. — Reproducibility Project

The Reproducibility Project did not follow the methodology of the original studies, but created new methodologies allegedly suitable for confidence in reproduction. What effect does the different approaches have on the outcome of the studies? How much of the inability to reproduce results hinges on the changes?

I have not read the report itself, only reviews by media and summary by UOV said about the study. The NY Times indicated that 250 researchers choose 100 experiments conducted in 2008 to test and they contacted the original experimenter to try to mitigate against the issues you suggest, but that was not always possible.

Statistically if 10% of the studies overstated results that would be one thing, but for 62% of the studies to overstate results is quite another matter and very much a concern.. Also note that this issue may not be confined to Psychology, all the life sciences are under hyper-pressure to produce results that can be published.

I understand the pressure to produce results can lead researchers to "fudging" results or building biases in to the studies, but the Reproducibility Project also strikes me as prone to such criticism. As you stated, 62% is significant. Is it possible that such a high number is the result of researchers in the project looking particularly hard for deviation rather than reproduction to spice up the findings? A study of agreement is far less sexy than a study of, "oh my God, our discipline has problems!"

Some thoughts.

Yes, these reviewers may have a built in bias, but I have not seen (but I will take a look) any substantive rebuttals.

Psychology experiments are inherently murky (to appropriate a term), which might change the manner of investigation, and how research reports are interpreted. Peer review might not be sufficient to confirm or deny results and duplication of results might be necessary for acceptance of any results.

"Measure twice, cut once". Old seamstress adage.

Baleeted.

For most sciences, one can take it for granted that the subject matter is unaffected by the theories one has about it. This does not at all hold for psychology; one's psychological understanding is an integral part of one's psychology. This is why most psychological experiments involve deception.

It also explains why psychological theories run in fashions, because as a theory becomes popular, it changes the psychology of the public to the extent of becoming unreliable. Thus for examples, Freudian theory has changed the general attitude to matters sexual, and notions of body language have moved body language itself from being a largely unconscious 'tell' to being a conscious and practiced 'display'.

One might say that publication ruins most psychological theories.

Interesting.

Psychology covers a lot of ground. It includes studies of eye movement (which might slop over into the terrain of neurology and ophthalmology), intelligence, learning, perception, memory, concentration, consciousness, personality theory (which might slop over into sociology, neurology, philosophy, pharmacology, etc.), way finding, cognitive processes (which definitely slops over into neurology), feelings, emotions, social behavior (which definitely slops over into sociology), and so on. It's a sloppy field.

That said, some fields of psychology have better success than others: learning, memory, and perception deal with observable and precisely measurable behavior. This area of study has been going on for over a century, and (as far as I know) the results hold up pretty well. For instance, psychologists have shown that a variable schedule of rewards yields longer lasting behavior changes than fixed reward schedules. This finding (made quite some time ago) explains why slot machines are so addictive: the slots reward players on a variable schedule.

Psychologists have done a good job, so I believe, of studying the bare facts of perception. How does the eye, ear, nose, and touch work? What can people actually perceive, and what can they not perceive? (Psychological studies have shown that people who read braille can also have dyslexia. Neurology will explain how dyslexia occurs in the brain.)

Studies of perception shape the way advertising is constructed. I'm not talking about speculative topics like "is the shampoo bottle phallic or not" but rather, how the eye moves over a page or a screen, and where the most important information (like brand name) should be placed for most certain perception.

Psychology makes the most progress when it can use animals or direct human observation. It's possible to study animal behavior in a lab in ways that most humans just will just not sit still for. Fortunately, a lot of what rats do also applies to humans. Neurons fasten on to the connection between action and reward regardless of whether they are in a rat's brain or a human subject's brain.

Rats, however, have rather simpler personalities and social lives than humans do, so studying rat personality doesn't get us very far in understanding human personality. Rats grow up and die of old age fairly quickly. A human will normally outlive most researchers funding, several times over.

From what i have read, there is also quite a bit of schlock being passed off as research in the social sciences. Almost any study that relies on surveys as a source of data is suspect, especially if it is retrospective. Suppose you want to study how much time students spend studying. "During the previous six months, how much time did you spend studying history?" Would that number be worth anything? Almost certainly not. Would it help if students logged their time? (phones can be used for this purpose)? Yes. It would be more valuable. If there was some way of checking on self reports it would be more valuable. Like, does geolocation information show that they are in the library when they say they are in the library? Are they studying history on line when they say they are on line, or are they playing games? The most useful data would come from observing students in some sort of unobtrusive controlled environment. That, however, costs money.

From what i have read, there is also quite a bit of schlock being passed off as research in the social sciences. Almost any study that relies on surveys as a source of data is suspect, especially if it is retrospective. Suppose you want to study how much time students spend studying. "During the previous six months, how much time did you spend studying history?" Would that number be worth anything? Almost certainly not. Would it help if students logged their time? (phones can be used for this purpose)? Yes. It would be more valuable. If there was some way of checking on self reports it would be more valuable. Like, does geolocation information show that they are in the library when they say they are in the library? Are they studying history on line when they say they are on line, or are they playing games? The most useful data would come from observing students in some sort of unobtrusive controlled environment. That, however, costs money. — Bitter Crank

Interesting points. I think it's just that some of the "discoveries" of psychological research don't seem to carry the same baggage as say, the applications of a protein interaction in regards to human biology. I suggested it here in that realists often rely heavily on brain mechanisms, but if they posit a grand theory for how conceptual thought is rooted in biology, it is going to have a tough time substantiating much when that "bridge" of neuroscience and psychology occurs and the the efficacy and content of research on the psychology side doesn't necessarily coincide with the research on the neuroscience side.

Also, your point about surveys seems justified. This goes back to my arguments against how people evaluate their lives. Yes, most people may claim their life is good or they are happy in a survey, but this is not necessarily the case when people are "in the moment" of everyday living in which case a much more thorough analysis is needed. These kind of surveys may be subject to the Pollyanna principle whereby one's evaluation of an event becomes positive even if during the moment of the event, it was not positive. This is also not to mention social pressures to conform to the idea that one must say a certain thing a certain way if one is to properly conduct themselves in society.

Any examples of papers with particularly bad methodology?

I bring this up because in a roundabout way, this problem of psychology's impotency to be used as "real" evidence will be hard for any realist account of concepts or meaning. A very fanciful and elaborate construct of how meaning is imputed through physical forces can be explained, but never verified via science due to there being no fitting psychological experiment that would definitely be agreed upon as truth to that elaborate theory. — schopenhauer1

Thanks for tagging me!

There are a number of interrelated issues here. I think the problem you indicate is real, but that it doesn't really turn on the reliability of psychological research. The trouble is, rather, that psychological research (for lack of any better alternative, after internal problems rendered the behaviorist program untenable) more often than not relies on an unreduced commonsense mentalistic vocabulary ("belief," "impulse," "appraisal," "attitude," etc) that doesn't admit of the kind of law-like generalizations we seek in the physical sciences. As Davidson and many others have argued, we simply use our psychological vocabulary in a very different way than we use our physical vocabulary.

However, if we take Sellars seriously on this front, we should understand even the fundamental psychological notions ("belief," "desire," "impression," etc) as having an essentially explanatory significance: we can conceive of them as theoretical postulates--unobservables--which serve to explain observable human behavior. In acquiring the psychological vocabulary, we learn to apply bits of it not only to explain the behavior of others, but also to recognize certain of our own experiences as instantiating those bits (e.g. recognize an experience as instantiating the desire to sit down). Understood in this sense, our psychological vocabulary does very good explanatory work.

That said, what is most explanitorily powerful in our practical lives (folk psychology), may not be what is most explanitorily powerful in giving a comprehensive picture of how the world hangs together metaphysically. It's perfectly plausible that we will continue to employ folk psychological notions in our practical lives, and even depend on them to make sense of our scientific ambitions, while simultaneously offering alternative theories of cognition, conation, etc, which are more amenable to law-like generalization. There is good reason to believe that such explanatory frameworks would have little role to play in our day to day lives, but they would certainly allow for more effective explanations of (for instance) the material conditioning of language. In fact, this is precisely the alternative approach to the issue we see on evidence in neuoroscientific approaches to ostensibly psychological questions, and precisely the application to semantic concerns we see in the field of neurolinguistics.

It's perfectly plausible that we will continue to employ folk psychological notions in our practical lives, and even depend on them to make sense of our scientific ambitions, while simultaneously offering alternative theories of cognition, conation, etc, which are more amenable to law-like generalization. — Glahn

The problem I see here is that terms such as "cognition" and "conation" and the like are as much folk psychological notions (despite being more 'technical sounding' and not as apparently explanatorily efficacious) as "belief, "desire" and "impression", and I cannot see how the former could be any more "amenable to law-like generalization" than the latter.

The problem I see here is that terms such as "cognition" and "conation" and the like are as much folk psychological notions (despite being more 'technical sounding' and not as apparently explanatorily efficacious) as "belief, "desire" and "impression", and I cannot see how the former could be any more "amenable to law-like generalization" than the latter. — John

This seems right to me. I should have been more cautious with my language! Speaking of "alternative theories of cognition," etc is not the right way to couch this issue.

What I should have said is this: though our folk psychological vocabulary has a great deal of explanatory efficacy in a number of domains, it is not of much use in answering questions like "How does the physical environment condition the structure of language?" My proposal is that we identify counterpart concepts in the physical vocabulary, which do allow for answers to these kinds of questions, allowing for the following concessions:

1. Our ability to recognize a particular concept as a counterpart to our folk psychological notions of cognition or conation quite obviously presupposes the resources of our folk psychological vocabulary. The latter is epistemically primary with respect to the physicalist counterpart vocabulary.

2. The counterpart concept of (say) cognition will not capture the essentially first-personal aspects of the concept as it figures in ordinary discourse, just as counterpart talk of norms as descriptive regularities loses the essentially first-personal rulishness of them.

It can be readily admitted that such counterpart concepts do not capture the depth of their folk psychological models, but also maintained that the counterpart concepts better afford answers to questions about (for instance) the material conditioning of the structure of language.

Yes, I definitely agree with you that "counterpart concepts" may be more amenable to causal accounts in physicalist terms, and may allow for some more intelligible crossovers between "the space of reasons" and the "space of causes", to put it in Sellar's terms.

This seems to be the age-old problem of understanding exactly where the humanities meet the sciences, with the more "human" sciences such as economics or history using, to varying degrees, mixtures of a language of reasons and a language of causes.

I do think it is important to develop as much as possible ways of giving account of all aspects of human life in fully naturalized terms.

Excellent post.

I dont think we can pick any one science as the only one relevant in any case. There is different studies for different things. Psychology only studies outward behavior. Biology and Chemistry study something different. However with each field you must learn it very well and use it properly to have any successful responses. If you are lazy you come to faulty conclusions. That does not discredit the field so much as show the stupidity of the person trying to make any claims.

I don’t think I have anything original to say about the philosophy around psychology, but I do have a unique perspective. I design tests to measure advertising. Everyone on my team has a copy of Experimental Design: Procedures for the Behavioral Sciences , by Roger Kirk. For anyone who doesn’t know, ‘Design of Experiments’ (DoE) is a topic-neutral synonym for ‘scientific method’. Many of the leading DoE experts are psychologists precisely because behavior is the trickiest terrain in the eyes of science. Ironically, there’s a (weak) sense in which psychology is the most scientific science.

That said, DoE got me interested in psychology, and I’ve felt increasingly disappointed following the rise of cognitivism. One hopes to exchange Skinner for some kind of methodological behaviorism, but now we’re inventing all these “systems” or hypothesizing that love can cause a starved cat to ignore food for its owner’s affection. Much psychology is just glorified market research.