Perhaps academia finds happiness in deconstructing stereotypes. In an age of automatic transmissions and self-driving cars it is a noble effort. Disambiguating the several meanings of the term stereotype in order to be referencing a similar meaning probably is worthwhile generally. I suppose Professor Bloom sees a linear growth of subjective and collective social epistemology because of his work in cognitive progress for infants to adults. Professor Bloom took up some very old philosophical studies regarding words and objects reference relations and lightly went over the way percepts are categorized and compared to other concepts for purpose of classification, labeling and so forth. At some point that process gives rise to explicit and implicit stereotypes or stored judgments in mind. I am not content to let the notion go unchallenged at least without making a couple of remarks about the use of the word as a pejorative use-truth that presumptively need be corrected.
Human minds are not quite computers with stored data from sensory input. Yet human thought does give the ability to form clusters of associated percepts and form meaningful neural pathways and iterations among them, without arbitrary limits. Stereotypes are incomplete information in a world experience of incomplete information. would point out the usefulness of stereotypes for sailors in dealing with weather and coastal phenomena regarding rocks, sand. fogs and so forth that allows one to evaluate partial information and make a judgment regarding a stereotype recurring in real time. Even such stereotypes or information regarding the draw down of water on a beach might give one time to flee from an approaching tsunami.
Logic structures regarding though classification from percepts and abstract data from studies provide a data base for individuals from which they may form opinions. Opinions different from true knowledge as Socrates pointed out so long ago, yet opinions are often the best one has. Incompleteness theorems abound in human experience because of temporal decay of any complete complex of compresence. Past, present and future along with position are never still. Accurate sorting and classifying of data regarding input with speed allows one to win at blitz chess and comprehend given worldly circumstances as they occur. Sterotypical patterns of behavior, presentation, or meaning of objects where they occur are useful tools for addressing empirical challenges.
It might not be too scientific to place moral values with economic values and cognitive or implicit epistemological mind construction patterns and neural virtual wiring. For me, moral questions should be something else. The concern about stereotypes psychologically speaking might rightly be considered as learning and epistemological concerns instead of morality.
Perhaps psychology has reacted differently to words and objects (W.V.O. Quine) differently than some philosophers in the post Two Dogmas of Empiricism era of linguistic philosophy regarding meanings, language and how they relate to 'objects'. Philosophers realize that words and objects are constructions rather than things in-themselves, although there are consistent patterns of physical interpretation of sensory input from the heterodox physical field that we regard as the quantum Universe field.
Stanford's Encyclopedia of Philosophy has an interesting article on social epistemology. https://plato.stanford.edu/entries/epistemology-social/
Social epistemology is different than individual subjective epistemology (it is more modern anyway). Some words and meaning values are socially defined- at least the lexicon of a group is from a social context. I just wanted to point out than anyone can use some practice on improving their personal epistemological tools. Stereotypes are a kind of higher level programming language of phrases that an be very useful when not incorrect. They may also be upgraded with better knowledge. Dumbing things down isn't necessarily the best option.
Epistemological limits to what anyone knows about anything including themselves should prompt some self-skepticism regarding the capacity for certainty about anything. Yet one should be able to have something like a 90% probability of accuracy, at least, in having opinions about ordinary matters of experience. Inaccuracy, falsehoods and so forth can be deceptive for-oneself as well as for-others.
I think its wrong to conflate cognitive associative tendencies regarding interpretation of sensed, input semiotic objects/patterns with morals and economic questions of value. People choose choose things on the basis of value to-themselves rather than making choices solely because of archetypal or stereotyped prioritization.