2 ms·

> It's not unethical to try to influence someone, even subconsciously, even by the choice of arbitrary design elements ("you start with 2 of 12 checked") -- as

by jbm 14y ago

> It's not unethical to try to influence someone, even subconsciously, even by the choice of arbitrary design elements ("you start with 2 of 12 checked") -- as long as there's no deception I feel your statement is too pat. Why is that not unethical? Based on what ethical pattern? If the basis of ethics is being truthful, it would seem that ease of understanding is also an important part of ethics. Subconscious manipulation and influence seems to be at odds with that goal. Perhaps such manipulations is inevitable in a marketplace where everyone engages in unethical mental manipulation, but that does not make it ethical.

gojomo 14y ago
But do we even know the 2/12 variant is worse on "ease of understanding"? Maybe it's more understandable, and congruent with the buyer's real best interest, because it's not subject to an endpoint/inertia-of-zero bias that depresses sales from the 0/10 case. Note that these same sorts of perceptual influences are also the basis for all the do-gooder initiatives that get proposed by Richard Thaler and Cass Sunstein in Nudge. (So in fact I would grant there is at least one more dimension beyond deception: is the influence clearly against someone's interest? But there should be no presumption that whatever they would have done without the influence was the purest expression of their interests.) Part of my ethical criteria is: what rules work? Part of 'work' is that they are evaluable. An accusation of undue influence, even though every observable aspect of a communication is true and fully disclosed, is very hard to evaluate. It's not very useful as a rule for deciding what one can do, or for assessing 'guilt' after the fact. Using a bit of communication that is false on its face is easy to evaluate. It's fraud, and it's clearly unethical. 'Manipulation' is a loaded term. It just means "influence we don't like". In a commercial setting, reasonable people already assume a seller is crafting their communication to increase sales. Calling cleverly-crafted commercial influence 'unethical manipulation', even when it remains true in all observable aspects, overloads and dilutes the idea of what's ethical. Is A/B testing that results in color, wording, and layout changes that increase sales 'unethical manipulation'? Perhaps if it goes so far, in total effect, as to create a tangibly false idea in the customer's minds. But not simply because it incrementally nudges someone towards a purchase.

by jbm 14y ago

gojomo 14y ago

But do we even know the 2/12 variant is worse on "ease of understanding"? Maybe it's more understandable, and congruent with the buyer's real best interest, because it's not subject to an endpoint/inertia-of-zero bias that depresses sales from the 0/10 case. Note that these same sorts of perceptual influences are also the basis for all the do-gooder initiatives that get proposed by Richard Thaler and Cass Sunstein in Nudge. (So in fact I would grant there is at least one more dimension beyond deception: is the influence clearly against someone's interest? But there should be no presumption that whatever they would have done without the influence was the purest expression of their interests.) Part of my ethical criteria is: what rules work? Part of 'work' is that they are evaluable. An accusation of undue influence, even though every observable aspect of a communication is true and fully disclosed, is very hard to evaluate. It's not very useful as a rule for deciding what one can do, or for assessing 'guilt' after the fact. Using a bit of communication that is false on its face is easy to evaluate. It's fraud, and it's clearly unethical. 'Manipulation' is a loaded term. It just means "influence we don't like". In a commercial setting, reasonable people already assume a seller is crafting their communication to increase sales. Calling cleverly-crafted commercial influence 'unethical manipulation', even when it remains true in all observable aspects, overloads and dilutes the idea of what's ethical. Is A/B testing that results in color, wording, and layout changes that increase sales 'unethical manipulation'? Perhaps if it goes so far, in total effect, as to create a tangibly false idea in the customer's minds. But not simply because it incrementally nudges someone towards a purchase.