I have unfounded suspicions that cuttlefish and squid, being more visual hunters, would be more enriched by visual rather than tactile stimulation, but it's hard to say, even for more "well understood" animals, what is "enriching." I've never been clear if cats are really "enriched" by strings, laserpointers, or catnip, exactly... a lot of people do seem to equate "enrichment" with "things people think are cure" or "things people associate with things that make kids happy." Despite that cynicism, I seem to remember that there are studies that show that some zoo animals have depression or lethargy sorts of problems that can be addressed by enrichment.
One of the things I wonder a lot about when reading the anecdotes around here is how much we anthropomorphise cephs we interact with... there's some evolutionary reason to believe that all mammals share some neurobiology and hormonal traits that means that assessing happiness, depression, anger, fear, comfort and such in cats, dogs, whales, pigs, rats, and so forth has some validity (although even there, there's probably a danger of humanizing them too much.) But in the case of cephalopods, there's no direct lineage or similarity in nervous systems that would make one have any reason to know how they think-- any evolutionary link goes back to something like a snail...
People, of course, are very good (and often unconscious) about using their human empathy social skills to predict, assess, and judge what something is thinking (I bet most of us even do that with our computers... my computers tend to be sadistic and enjoy torturing me.) But even correcting for that, I wonder if the observations of cephalopods showing behaviors like moping, or anticipating, or anger, or excitement, or play show that these are emergent behaviors of any intelligent animal, or any intelligent animal based on the metazoan body plan and physiology, since they appear to have evolved similarly in a separate lineage.
Unfortunately, I don't have any better suggestions for measuring enrichment, but it does seem like, at least in many anecdotes, people can learn to read ceph body language and behavior in a way that is connected with their demonstrated active choices: Carol's octos seem to actively prefer to manipulate the green lego, and some octos actively avoid or hide in the presence of light, or camera flashes, or humans, or fish. And it seems like there are examples of octos actively calling attention to themselves when they're "bored" although it's hard to prove how much that's related to their learning that the humans provide food. But there are certainly a lot of anecdotes about cephs developing habitual behaviors that don't seem to be directly related to feeding... I can think of a few reports of divers interacting with cephs in the wild that suggest curiosity and socialized behavior more than just feeding-related attitudes, although it's always hard to tell for sure if the base curiosity is driven by the "can I eat it?" and "can I mate with it?" questions (although some human psychologists might argue that that's all there is to human behavior, as well.)