Clever Hans

[nullve]

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

[mgoetze]

?!?

[nullve]

?!?

Consider how Hans and his owner were unwittingly using a code to "cheat" at arithmetic.

[helene_t]

Ah right, fiddling with the bidding cards. Touch all of them, starting with pass though 7NT, as soon as you partner shows tension you know he doesn't want you to fiddle any more.

[Zelandakh]

This happens at club level. Everyone must surely know at least one player where the whole table can immediately see (s)he has a big hand.

[ggwhiz]

This happens at club level. Everyone must surely know at least one player where the whole table can immediately see (s)he has a big hand.

 

We had a player who moved her lips while counting her hcp's. My partner in 3rd with her in 4th once psyched when she got to 20 and was still counting.

[nullve]

"Years ago I took to filming my players as they trained for competition in order to ascertain what tells they owned. One certainty became apparent virtually immediately: every single player possessed at least one and typically three or more. The worst ones I revealed to them, primarily because they were placing themselves at an extreme competitive disadvantage. I also became concerned that their partners would subconsciously learn them, too." (http://bridgewinners.com/article/view/the-unintentional-and-its-offspring-part-1/)

[nullve]

Do the von Ostens and Clever Hanses of bridge actually cheat? If not, how can we as spectators tell the difference between them and real cheaters?

[Zelandakh]

Do the von Ostens and Clever Hanses of bridge actually cheat? If not, how can we as spectators tell the difference between them and real cheaters?

Almost all "tells" of this type are going to be invisible to partner in a screens environment. Generally, cheating across a screen is more difficult to pass off as incidental once the key is known. Given your exercise, perhaps you would have been better off videoing the opponents and finding out their tells!

[nullve]

Almost all "tells" of this type are going to be invisible to partner in a screens environment.

Maybe. But:

 

"During a practice match between Junior Teams conducted over twenty years ago, I witnessed a new player placing his opening bid in the center of his space when he held an average hand, to the far left with a great hand and somewhere between those physical points with an intermediate hand. Given that he had only been playing for a few months and sitting in a new partnership, it was obvious that this was not an effort to cheat. It was simply that he was unconsciously leaving room for the bidding cards to be comfortably placed for the expected duration of the auction." (http://bridgewinners.com/article/view/the-unintentional-and-its-offspring-part-1/)

 

On the face of it, this is not very different from what B-Z allegedly were doing.

[nullve]

From a comment on Bridgewinners:

 

"If someone is cheating, do we have to disclose their full method to prove guilty? No, we don't.Since B-Z use 3 ways to bid(small, normal, large gaps), so if they are innocent they are doing it unconsciously, and it will be quite random. If evidence show that randomness is violated then it is serious, regardless of the exact meaning." (http://bridgewinners.com/article/view/a-different-approach-on-b-z-bidding-gap-issue-3/)

 

This got me thinking about the game of rock-paper-scissors:

 

"Proponents of the “Chaos School” of RPS try to select a throw randomly. An opponent cannot know what you do not know yourself. In theory, the only way to defeat a random throw is with another random throw – and then only thirty-three percent of the time. Critics of this strategy insist that there is no such thing as a random throw. Human beings will always use some impulse or inclination to choose a throw, and will therefore settle into unconscious but nonetheless predicable patterns. The Chaos School has been dwindling in recent years as tournament statistics show the greater effectiveness of other strategies." (http://worldrps.com/advanced-rps/)

[nullve]

The hidden game of "rock-paper-scissors" in a game of bridge with bidding boxes:

 

Players: the two members of a bridge partnership

 

Moves: gaps between consecutive bids by one player that are either

 

* 'small' (plays the role of, say, 'rock')

* 'mid-sized' (plays the role of, say, 'paper')

* 'large' (plays the role of, say, 'scissors')

 

Since the players take turns to bid and thereby make a physical move in this game, it's not at first sight a simultaneous game like "real" rock-paper-scissors. But as long as the players focus on the bids instead of the gaps, the last player to move will have no advantage except possibly due to subconscious effects. So at least we can pretend it's a simultaneous game, albeit one played unwittingly.

[1eyedjack]

* 'small' (plays the role of, say, 'rock')

* 'mid-sized' (plays the role of, say, 'paper')

* 'large' (plays the role of, say, 'scissors')

* VERY 'small' = Lizard

* VERY 'large' = Spock?

[nullve]

* VERY 'small' = Lizard

* VERY 'large' = Spock?

Yes, why not.

[nige1]

Some declarers peter from equals in dummy, to reassure dummy that the contract is in the bag. Seemingly harmless but might be illegal, according to the laws.

[helene_t]

Some declarers peter from equals in dummy, to reassure dummy that the contract is in the bag. Seemingly harmless but might be illegal, according to the laws.

I usually apply the same rules for adjacent honours when playing from dummy as when defending. For no particular reason. As a child I played the lower because it would do more harm if I forgot whether a high honour had been played :)

 

I don't think the laws make any restrictions on the order in which declarer plays equivalent cards. Maybe some would say that playing the cards in a weird order, to cause distraction, is in the same ballpark as a colour coup.

[gwnn]

Can anyone explain how Clever Hans relates to the codes we are talking about? Or RPS? As if I were a five year old? I understand that the owner's unconscious signals told the horse when to neigh, and I'm sure some people do things like when they're defending/bidding (for example, "please pass.. Please pass.. My raise was a joke, please don't double them..." or during defense, "DON'T CASH THAT ACE! It's a ruff/sluff") and I even worry that I'm doing something along those lines while I'm playing live. But we're talking about a partner signalling distribution by showing fingers, or talking about placing vertical/horizontal cards when you are attacking/staying passive. This is completely different to trying to manipulate your partner to do something (unconsciously). And the RPS example, I have no idea what possible relation it could have to sending signals. The RPS players are actively trying to predict how their opponents are trying to behave. In bridge, you're supposed to ignore everything your partner sends and perhaps play some RPS with your opponents (although not in the realm of "tells" as it is illegal to mislead your opponents through mannerisms). But I guess I'm missing the point since several posters seem to think the OP is onto something.

[nullve]

Can anyone explain how Clever Hans relates to the codes we are talking about?

 

Well, everyone seems to assume that in order to prove collusion, it's sufficient to prove that an illegal signal has been transmitted and acted upon. I'll call this view 'transmissionism'. The Clever Hans case, however, indicates that an illegal signal can be transmitted and acted upon without collusion, because, apparently, von Osten didn't know he was signalling and Hans wasn't aware he was doing artihmetic. In bridge, a von Osten-Clever Hans-type pair would be a pair A-B where

 

A had tells he doesn't know about

B was using those tells subconsciously

 

Would A-B be cheating? If transmissionism is true, they would.

 

Do von Osten-Clever Hans-like pairs actually exist? For all I know, they may be rule rather than the exception.

 

Or RPS?

The game of RPS suggests that we are all von Ostens, because it's notoriously difficult to play a long sequence of moves in RPS that doesn't reveal a pattern that can be exploited by the opponent. In fact, humans suck at RPS compared with certain computer programs.

 

But we're talking about a partner signalling distribution by showing fingers, or talking about placing vertical/horizontal cards when you are attacking/staying passive.

 

How do we know these signals aren't tells? I'm not saying they are, but a proof of collusion should be based on more than a gut feeling that they are part of a

preagreed code.

 

This is completely different to trying to manipulate your partner to do something (unconsciously).

 

I'm not sure what you mean. von Osten apparently didn't try to manipulate Hans, he just did. And even if we assume that F-N and B-Z were cheating, it doesn't have to be the case that those hand movements or card orientations were deliberate as long as partner knew about them and took advantage.

 

And the RPS example, I have no idea what possible relation it could have to sending signals.

 

I was trying to rebut this:

 

"If someone is cheating, do we have to disclose their full method to prove guilty? No, we don't.Since B-Z use 3 ways to bid(small, normal, large gaps), so if they are innocent they are doing it unconsciously, and it will be quite random. If evidence show that randomness is violated then it is serious, regardless of the exact meaning." (http://bridgewinners.com/article/view/a-different-approach-on-b-z-bidding-gap-issue-3/)

[gwnn]

Really? Putting the board on the other side of the screen to signal for a club lead can be an unconscious tell? And how will partner possibly understand it? There's a reason for my examples. I am acting tense to get partner not to double, and the moment he doesn't double, I am relaxed. He will pick this up, either immediately or by training. He, being a decent human being, will sense my suffering and try to alleviate by passing. This is a totally human interaction, albeit a blatantly illegal one, based on unconscious tells that are natural to humans and empathy. Say I am in a sticky situation at the police station as I'm unsure whether I have all the necessary papers and I'm anxious whether or not they will grant me some permit. I will be naturally tense and hoping they just give it to me. When I preempt on less than a normal minimum and LHO doubles, I'm anxious RHO doesn't pass it out. This may or may not show, and RHO may or may not notice it (this time, using it is legal). These are all completely normal human reactions to common situations and I would argue, picking them up is not only natural but failing to pick them up could possibly be a sign that I'm a psychopath. All of this is completely different than placing a card vertically and horizontally, which maps nowhere on these basic human non-verbal interactions, or putting a board somewhere based on which suit is strongest (but only when you are defending), or clearly showing four or five fingers under the screen aperture. Can't you see how different those signals are to these?

[Zelandakh]

All of this is completely different than placing a card vertically and horizontally, which maps nowhere on these basic human non-verbal interactions,

While I agree wholeheartedly with the point you are making, I do take a small issue with this part. One of the most common such actions you see at club level is placing a card vertically when expecting to win the trick and horizontally when not. I doubt many of the players that do this are doing it consciously.

[nullve]

Putting the board on the other side of the screen to signal for a club lead can be an unconscious tell?

 

You're using intentional language here, already suggesting that the board was put on the other side of the screen deliberately, i.e. that it couldn't possibly be just a tell. But maybe your point is, as your examples suggest, that tells in bridge always consist of rather inconspicuous and involuntary motion, like nervous shaking. Then I want to remind you that in poker, very conspicuous and seemingly deliberate behaviour like drinking from a bottle can also be a reliable tell. So why not in bridge?

 

And how will partner possibly understand it?

 

In a true von Osten-Clever Hans-like partnership, there is no understanding at all! But the Clever Hans-like player will have learnt to use the tells to his advantage, maybe through something like operant conditioning.

[barmar]

While almost anything is possible (there's lots we still don't know about human psychology), I suspect that the signals that relate to specific suits are more likely to be deliberate. It seems like subconscious thoughts would more likely be about qualitative aspects ("good hand" versus "bad hand") than specific quantitative features (e.g. the number of hearts). Subconscious thoughts and actions are also often primed by context -- a tell might be related to the most recent suit bid, rather than always the same specific suit.

 

And we should also consider how "natural" the physical actions are. The Reese/Shapiro finger signals take some effort to pull of, and they were way too strongly correlated with a specific suit to be unconscious.

[gwnn]

Barmar explains it already very well but since I can't upvote him I need to reply as well.

 

Again, there is a very big difference between the following two situations:

 

a) It goes (in perfectly normal tempo)

1♦-p-1♥-p

4♥-4♠-5♥-p

p-???

I'm sitting there, slowly growing irritated, partly due to partner's "imaginative" bidding (which usually shows a lot of spades and a self-deception about his ability to "get more information by passing), and partly that apparently he doesn't care about my decision to pass over 5♥. I'm sitting, waiting, knowing that the more time passes the less likely partner is going to pass out 5♥ (I have something like an 1354 4-count so would really prefer partner to pass. we may even set 5♥). I'm getting pissed off and doing my best not to show it. I don't know how well I manage. {Eventually, as you may imagine, he does bid 5♠ and goes for 500 against -420 at matchpoints, and bemoans that he "couldn't have known whether I have a singleton or a trebleton." Well, at least we are not cheating.}

b) I have AK of hearts and I put down the board on the side of the aperture on the screen. I continuously do this and arrange the board when I see that the board is not at the side enough.

 

There might be middling cases, such as "I led a singleton, my partner won the ace and I'm waiting for his return. I don't understand what he's thinking about and I really just want him to do the <<normal>> thing, i.e., returning my lead." This is perhaps more related to the case of "specific holding in a suit" but this is still a question of me wanting to have partner do something that is somehow the default action in a hand.

 

So maybe that is what I am saying: is there a "default action I expect partner to do?"

 

In the case of Clever Hans, yes, there is. I want him to neigh at a specific time. I know that the owner did not think he was giving Hans this information, and he did not think he was passing a message, but there is no doubt that the owner did want to see him neigh at a particular moment (much like I want my daughter to do well in gymnastics, I am eager to see her succeed). When partner is sitting there thinking what to do, clearly there are only two (or three) options available, and the default one is pass. I just want him to do what he's supposed to (i.e., not bid his hand twice) and maybe I give away this eagerness. But this is still not a very well encoded message about a particular suit or who knows what.

 

Of course, it is possible that there is some sort of evolutionary mechanism where I scratch my head every time I have 5 hearts, and I play with a lot of bridge with a lot of partners, and the ones who unconsciously always play me for 5 hearts when I scratch my head unconsciously, I score a tiny bit better with those partners so I will tend to those partners, and so on. But do you really think this is a plausible scenario? We can theoretize all we want, and I cannot prove to you logically that B/Z aren't the people who show 5 fingers whenever the number "5" enters in their minds, but this doesn't quite add up.

[gwnn]

Furthermore you are neglecting the fact that in the case of BZ there are a few instances where they actually adjust the gaps between the bidding cards. It looks quite suspicious. Actually the "bidding gaps" theory is the only case in which I'd be a bit on board with the perhaps unconscious idea but the adjusting is pretty damning. Of course it could be the case that my unconscious mind tells me that the "proper place" of the cards is well-spaced, and this "proper place" that my unconscious mind suggests could change from board to board as a function of strength, and when the bidding card is not where its "proper place" is, I'd adjust consciously for no reason I could name, but can't you see how many ifs and weird suppositions I needed to make? I can rationalize anything you want, just tell me something and I can rationalize it, but some rationalizations are more rational than others.

[gwnn]

Stuff where I'd be on board with your unconscious/Clever Hans idea:

- some people play random spot cards messily and play signalling cards (the ones that actually mean something) very deliberately. This could just mean that they consider the signalling cards somehow "more valuable" so they handle them gently. Of course it could also mean that they are trying to get partner to notice the cards, which would be cheating, but I'm OK with giving them the benefit of the doubt.

- playing singletons could have a different hand motion. this is actually something I spend a fair bit of time worrying about (for myself). For example, picking out a singleton could be done a bit more carefully as missing a card could end in a revoke, while picking out a random spot from five is a more error-tolerant endeavor.

- many many other stuff

 

Showing 5 fingers to show 5 hearts? Very implausible. Putting the board on my side of the screen when I want a spade lead? No sorry I don't buy it. I would not buy it even if you gave me free money with it.