English Base

Test 019

Reading Practice: Colour Words and Perception

Reading Text

Set two squares of blue side by side, one deep cobalt, the other the pale blue of a winter sky, and ask a Russian speaker what they see. The answer comes without hesitation: two colours, siniy and goluboy. An English speaker, shown the same pair, is likelier to report one colour in two shades, a difference of degree. Neither observer is lying. The question is whether the words they reach for merely describe the difference or quietly help to produce it.

The intuition that language might shape perception has a long, uneven history. Nineteenth-century philologists read vocabulary as a window onto the mind of a people; the twentieth century, recoiling from such claims, settled for something more modest and testable. Berlin and Kay's survey of basic colour terms held that languages acquire their colour words in a rough sequence. It concerned universals rather than the effects of naming, yet left an awkward residue: if every language draws the same distinctions, why do the boundaries fall in such different places, and why does a boundary marked in one language go unmarked in another?

The most disciplined attempt has come from laboratory work on categorical perception. In one well-known design, Russian speakers saw three chips in a triangle and asked which lower chip matched the upper. When the odd chip out crossed the siniy–goluboy boundary, responses were faster than when all three belonged to one category. Crucially, the advantage vanished when participants rehearsed an unrelated string of digits while doing the task, a manipulation meant to occupy the verbal resources naming requires. The inference is elegant: if blocking the inner word erases the benefit, the word was doing some of the work.

Such findings are narrower than they sound. The effects last tens of milliseconds and depend on stimuli chosen to straddle a boundary known to be linguistically marked. Nor does a faster response settle what is perceived: a decision may be speeded by a label without the sensation itself being altered, much as a familiar face is recognised sooner without looking any different.

Memory tells a related story. Asked to pick out a colour seen moments earlier, people drift towards the best example of the category they named it with — prototype distortion. The label works after the fact, pulling a remembered shade towards the centre of its category. That the drift follows the naming, not the reverse, has been shown by manipulating the words available: offer a finer distinction and memories grow finer, if only slightly. Boundaries, then, are gradients rather than walls, and where a language places them is an accident its speakers inherit rather than choose.

None of this licenses the comfortable conclusion that speakers of different languages inhabit different perceptual worlds. The evidence points to something more restrained: colour categories are real enough to be measured, words sharpen attention at their edges, and the sharpening is partial, task-dependent and easily undone. What the two blues show is not that the eye is a prisoner of the dictionary, but that the dictionary is one of the instruments the eye has learned to use — and an instrument, unlike a prison, can be put down.

Comprehension Test

Question 1. The remark "Neither observer is lying" implies that

✗ IncorrectNo measurement of light is performed anywhere in the paragraph; the squares are identified by everyday names. The remark concerns the sincerity of the two descriptions, not their physical precision.

✓ CorrectThe Russian sees "two colours, siniy and goluboy", the English speaker "one colour in two shades"; both say what their vocabulary makes salient. The clause concedes sincerity on both sides before the closing question of whether words "help to produce it".

✗ IncorrectThis reverses the phrase, whose purpose is to rule out exaggeration on either side. If one speaker were overstating the contrast, there would be nothing left to explain.

✗ IncorrectThat is precisely the position the passage sets out to test rather than assume. Were perceptions identical, the final question about words helping to produce the difference would be pointless.

Question 2. In saying the English speaker treats the pair as "a difference of degree", the author means that for this speaker the two blues

✗ IncorrectNothing in the sentence concerns instruments. "Degree" is contrasted with a difference in kind, not with the limits of the unaided eye.

✗ IncorrectThe English speaker does see two shades; what is at issue is whether they count as separate colours. The phrase describes classification, not the absence of a visible difference.

✗ IncorrectTaste plays no part in the paragraph, and the speaker is describing a colour, not expressing a preference. The contrast is between one continuum and two kinds.

✓ Correct"One colour in two shades" places the pair on one scale, as opposed to the Russian "two colours, siniy and goluboy", which the first paragraph pointedly does not call versions of one. That contrast sets up the question about naming.

Question 3. Which question does the opening paragraph leave the rest of the text to answer?

✓ CorrectThe last sentence of the paragraph states it almost verbatim: "whether the words they reach for merely describe the difference or quietly help to produce it". The survey, the experiments and the caveats that follow all bear on that alternative.

✗ IncorrectThe paragraph notes a single contrast between siniy and goluboy on one side and blue on the other, but never counts anything. A shortage of terms is not the issue raised.

✗ IncorrectThe text deliberately avoids optics and describes the squares by everyday names. Wavelength is never mentioned as a criterion.

✗ IncorrectTranslation is what makes the opening anecdote tellable at all, but it is not the puzzle posed. The closing question is about effects on perception, not about renderings between languages.

Question 4. The reference to nineteenth-century philologists chiefly serves to

✗ IncorrectThe view is reported and then left behind: "the twentieth century, recoiling from such claims, settled for something more modest". No endorsement is offered.

✗ IncorrectNo experiment is attributed to the philologists; they are placed a century before the laboratory work described later. Their method was the study of words, not of responses.

✓ CorrectThe sentence balances "read vocabulary as a window onto the mind of a people" against a century that "settled for something more modest and testable". The contrast prepares the ground for Berlin and Kay's survey.

✗ IncorrectThe surveys were not abandoned: Berlin and Kay's is introduced in the very next sentence as the modest successor to the older ambition. Its limitations are discussed later, but not its abandonment.

Question 5. What makes the "awkward residue" awkward?

✗ IncorrectTestability is not the complaint, and the survey is presented as the testable alternative to earlier speculation. The difficulty lies in the placement of boundaries, which the two questions spell out.

✓ CorrectThe residue is the pair of questions that follow: if every language "draws the same distinctions", why do the boundaries "fall in such different places", and why is one of them marked in one language and not in another? A universal inventory sits uneasily with variable boundaries.

✗ IncorrectNo such restriction is mentioned, and the survey's whole point is its breadth of coverage. The criticism concerns the theory's internal tension, not its sample.

✗ IncorrectNumbers of terms are not in dispute here. The passage fixes on where boundaries are drawn, not on how many categories a language recognises.

Question 6. By noting that the survey "concerned universals rather than the effects of naming", the author implies that it

✓ CorrectA survey of which categories languages possess leaves the causal question untouched, which is exactly why the laboratory work of the next paragraph is needed. Silence about effects is not a verdict on them.

✗ IncorrectThe author says the effects lay outside the survey's scope, not that they were ruled out. A denial would be a finding, and none is reported.

✗ IncorrectLiteracy is never mentioned. The distinction drawn in the sentence is between universals and the effects of naming, not between written and spoken languages.

✗ IncorrectThe later experiments qualify one another's claims; they do not overturn the inventory of basic terms. The author keeps the survey and adds a different kind of question.

Question 7. The triangle design was intended to measure

✗ IncorrectNo vocabulary count is reported. Participants chose among chips, and their inner naming was interfered with rather than counted.

✗ IncorrectParticipants were never asked to place the boundary; the experiment inferred its effect from how quickly they chose. Asking them to describe it would have been a different study.

✗ IncorrectAccuracy is not the measure reported. The comparison is between speeds when the odd chip did or did not cross the boundary, not between correct and incorrect matches.

✓ Correct"When the odd chip out crossed the siniy–goluboy boundary, responses were faster than when all three belonged to one category." Response time is the dependent measure, and category crossing is the variable.

Question 8. The rehearsal of digits was introduced in order to

✗ IncorrectA uniform slowing would leave the boundary advantage intact and explain nothing. The point was selective interference with one kind of processing.

✗ IncorrectThe task was a matching decision, not a memory test. Memory is discussed later and on quite different evidence.

✓ CorrectThe text calls it "a manipulation meant to occupy the verbal resources naming requires", and the advantage then vanished. That disappearance is what licenses the inference about the inner word.

✗ IncorrectConcealment is nowhere mentioned. The interference is described in terms of the resources it consumes, not in terms of what participants believed.

Question 9. The reasoning in "if blocking the inner word erases the benefit, the word was doing some of the work" is best characterised as

✗ Incorrect"Some of the work" is a deliberately modest formulation, and the next paragraph insists that the sensation itself need not be altered. A proof would leave no room for that qualification.

✓ CorrectThe sentence is conditional and holds only if the digits task occupied the relevant verbal process; had it interfered with something else, the conclusion would not follow. The author calls the inference "elegant" rather than conclusive.

✗ IncorrectIt asserts the contrary: remove the inner word and the benefit disappears, so inner speech matters. The option inverts the relation the sentence establishes.

✗ IncorrectNo such comparison is drawn, and the design described here does not include a second language group. The claim is about the role of naming in this task, not about who names inwardly.

Question 10. The author's principal reservation about these experiments is that

✗ IncorrectReplication is not discussed in this paragraph at all. The reservations stated concern the magnitude of the effects and the way the stimuli were selected.

✗ IncorrectThe author reports the study without comment on its sample and never treats a single-language group as a defect; the reservations actually stated concern the size of the effect and the choice of stimuli. A feature of the design is not thereby the author's objection to it.

✗ IncorrectThe author does not dismiss response times; the objection is narrower — a faster response does not by itself "settle what is perceived". A weaker version of the claim would be defensible, the absolute one is not.

✓ Correct"Tens of milliseconds", and dependent on "stimuli chosen to straddle a boundary known to be linguistically marked". Both details confine the finding to narrow, designed conditions.

Question 11. The comparison with a familiar face is introduced to

✓ Correct"A decision may be speeded by a label without the sensation itself being altered, much as a familiar face is recognised sooner without looking any different." The analogy separates the speed of a judgement from the character of the experience.

✗ IncorrectThe comparison is illustrative, offered "much as" another everyday case behaves. No claim about shared machinery is made or needed.

✗ IncorrectThe point runs the other way: recognition is quicker while the face looks no different. Vividness is exactly what the analogy denies is affected.

✗ IncorrectThe author questions one inference drawn from the tasks, not the value of experimentation. The experiments remain the backbone of the argument.

Question 12. Stimuli chosen to "straddle" a boundary are ones that

✗ IncorrectStraddling is the opposite of lying inside: the design requires items on both sides so that crossing can be compared with not crossing.

✗ IncorrectBrightness control is not mentioned; hue is what the boundary is drawn across. The word describes position relative to a division, not a photometric property.

✓ CorrectTo straddle is to sit astride, with some chips on one side and some on the other, so that a boundary effect can appear. The design presupposes a line the experimenter could locate in advance, and that presupposition is part of the author's objection.

✗ IncorrectPresentation time is not discussed. Naming is what the interference task suppressed, and only in the second condition.

Question 13. "Prototype distortion" names

✗ IncorrectNo new terms are coined in the passage. The phenomenon concerns what happens to a remembered shade, not the growth of a vocabulary.

✗ IncorrectFading would be loss of detail in any direction. What is described is a systematic shift towards one particular example, which loss alone cannot produce.

✗ IncorrectThe direction is reversed. The label is applied first, and the memory then shifts towards the best example of the category that label names.

✓ Correct"People drift towards the best example of the category they named it with — prototype distortion." The naming comes first in the sentence, and the pull towards the centre follows it.

Question 14. The author holds that the drift follows the naming rather than the reverse because

✗ IncorrectThat is far stronger than the evidence offered: the text shows only that a remembered shade gravitates towards the centre of the category it was named with, not that memories are verbal from the outset. The argument rests on manipulation, not on encoding.

✓ Correct"Offer a finer distinction and memories grow finer, if only slightly." Since the vocabulary was altered and the memory moved with it, the direction of influence is established by the manipulation.

✗ IncorrectDelay is not the argument; a delay by itself would say nothing about which came first. The evidence cited is the effect of varying the available words.

✗ IncorrectNo language restriction is claimed for the memory effect. Russian served the earlier discrimination experiments, and the passage does not generalise from that alone.

Question 15. "Boundaries, then, are gradients rather than walls" implies that

✗ IncorrectA gradient can still be perceived and marked by a word. The claim concerns the nature of the transition, not any blindness to it.

✗ IncorrectThe same sentence calls the placement "an accident its speakers inherit", which points to variation rather than agreement. Uniformity is what the paragraph denies.

✓ CorrectThe metaphor denies sharp edges, and the clause that follows calls the location an inherited accident rather than a choice. Both halves of the sentence have to be read together.

✗ IncorrectThe text says speakers inherit the placement "rather than choose" it, which excludes deliberate redrawing. Agency is precisely what the sentence removes.

Question 16. The final paragraph chiefly serves to

✓ CorrectIt opens by refusing "the comfortable conclusion" and then states what the evidence does establish: measurable categories and a sharpening that is "partial, task-dependent and easily undone". That is a qualification, not a summary of method.

✗ IncorrectNo procedure is described there; the paragraph works entirely at the level of interpretation. The triangle and the digits belong to the third paragraph.

✗ IncorrectNo new data appear; the paragraph draws consequences from material already presented. Its force is retrospective.

✗ IncorrectIt asserts the opposite: categories are "real enough to be measured". The scepticism is directed at a strong conclusion, not at measurement.

Question 17. The phrase "None of this licenses the comfortable conclusion" conveys

✗ Incorrect"None of this licenses" is a refusal, and "comfortable" marks the conclusion as tempting rather than established. The sentence opens the paragraph by withholding agreement.

✗ IncorrectFive paragraphs have been spent on that very question, and the paragraph goes on to specify what the evidence shows. The tone is engaged, not detached.

✗ IncorrectThe evidence is accepted as showing something; only the strong reading of it is refused. Dismissing the experiments would contradict the sentence that follows.

✓ CorrectThe refusal is followed at once by a concession: words "sharpen attention at their edges". Restraint about the conclusion coexists with a limited effect, which is the author's settled position.

Question 18. In the last sentence, the dictionary resembles an instrument in that it

✗ IncorrectThe sentence ends with the opposite image: "an instrument, unlike a prison, can be put down". Necessity is what the analogy is designed to deny.

✓ CorrectThe contrast is instrument against prison: a tool one uses and may set down, as opposed to a cell one cannot leave. This underwrites the modest conclusion the paragraph has just sketched.

✗ IncorrectNothing is said about who compiles dictionaries. The relevant property is that a tool is optional in use, not that it is professionally made.

✗ IncorrectObsolescence is never mentioned, and it would not explain the analogy. The dictionary here is something speakers actively use.

Question 19. Which best describes the overall movement of the argument?

✗ IncorrectThe experiments are treated as supporting a weak version of the thesis, not as hostile evidence to be resisted. Nothing in the text is defended against the data.

✗ IncorrectBerlin and Kay are introduced as a modest advance on an older ambition, not as an error to be corrected. Their work is kept, with a tension noted inside it.

✓ CorrectThe sequence runs: the two blues and the question; the history; the experiments; the reservations; memory; the qualified ending. Each stage narrows the claim rather than expanding it.

✗ IncorrectNo symmetry of support is claimed anywhere. The author comes down on the side of a partial effect, which is a verdict rather than a tie.

Question 20. Which claim would the author be most likely to accept?

✓ CorrectThis matches the closing paragraph: words "sharpen attention at their edges", and the sharpening is "partial, task-dependent and easily undone". It is the restrained thesis the whole text builds towards.

✗ IncorrectNothing supports a deficit claim: the English speaker sees the two shades perfectly well, and "one colour in two shades" is a classification, not a blindness. The text never counts what is perceived.

✗ IncorrectThe author expressly denies this inference: a decision may be speeded "without the sensation itself being altered". The familiar-face comparison exists to block exactly this reading.

✗ IncorrectThe text reports experiments on precisely these categories and calls them "real enough to be measured". Contingency of placement is not the same as untestability.
Previous Test 018 Next Test 020