Preliminary Analysis of Methods of Reading Rotated Text

Benjamin Fleischer

Undergraduate, University of Pennsylvania

The study was designed to determine the time to read spatially rotated text and thus begin research of the method by which it is recognized. Reading time of subjects was recorded with text at counterclockwise intervals of 45° with respect to the horizontal for 8 different paragraphs of meaningful text. Reading difficulty, as shown by increased mean times, peaked just at 135° and 225°. Readers then showed decreased mean time at 180°. This implies that a method faster than mental rotation is used. The exact method is beyond the scope of this paper.

In the development of reading machines, it is important for the machine to achieve a level of textual perception as flexible as a person's. "The visual nervous system possesses compensatory rectifying mechanisms by means of which it achieves 'constancy' of visual recognition" (Kolers & Perkins, 1975). An important rectifying mechanism used in everyday life is that of geometric rotation. Text is seldom found precisely parallel to the horizontal. Neurons that code for lines, edges, and orientations have been found in the nervous system (Sutherland, 1968 as cited in Kolers & Perkins, 1975). Thus, there is a physiological basis for this view. Further, Aulhorn (1948 as cited in Kolers & Perkins, 1975) found that reading times accommodate a cosine function as pages of a book were rotated through 360 with minima at 0 and 180 and maxima at 90 and 270. "Smith, Cambria, and Steffan(1964)… found that a rotation of 180° in the plane of the page induced less impairment on alphabetic text than did its constituent 180° rotations around the horizontal and vertical axes" (Kolers & Perkins, 1975). Kolers & Perkins (1975) stated that the above studies do not support a purely rotational component because of the asymmetry of rotation and reduced impairment at 180. They then went on to suggest three component of textual perception: a component that changes order, a component that flips characters horizontally, and a rotational component. They did not explain the decrease at 180. Changes in reading time are due to textual processing as semantic content remains the same (Kolers, 1973 as cited in Larsen, 1984). That a flipping component exists is supported by the rare case of dyslexic children able to read reversed or mirrored print better than ordinary material (Orton, 1937 as cited in Larsen, 1984). All in all, many studies have been done and the results are still not clear. The purpose of this study is to acquire new data and to attempt to begin rectification of the differences.

EXPERIMENT

Method

Subjects. Twenty-five undergraduate volunteers were tested individually. One subject was tested and replaced due to experimenter error and inability to successfully complete the task. Two others were thrown out for the same reasons.

Materials and design. A total of eight experimental paragraphs were found on the internet ([1,2,…8]) The paragraphs were all statistically similar (Table 1). Paragraphs were printed in True Type Times New Roman 11 point. There was no topical continuity among the pages.

Test. Each subject was told to read aloud at a normal, constant pace. Subjects were also advised to make a quick guess at any difficult words and move on so that any individual words would have no significant effect upon the entire reading time. The purpose of this was to reduce errors and measure solely the amount of time to read at the angles. Paragraphs were then given at counterclockwise angles of 0,45,90,135,180,225,270,315 degrees with respect to the horizontal. The order of the paragraphs was varied 12345678, 23456781,.. to 81234567 with the numbers always in order. The experimenter recorded reading times on a stopwatch. Time was paused during reading if a subject laughed or otherwise lost focus until the subject regained self-control. Subjects were prompted on words that they were having obvious difficulty on. If a line was skipped, the subject was asked to read the line.

Subjects. Successful participants were 22 undergraduates. Of these, 20 were right handed, 7 were male (both left-handers were female), and the first language for four were Hebrew, Portuguese, Korean, Chinese. Chou's (1929,1930) findings for Chinese logograms are similar to Aulhorn's (1948) (as cited in Kolers & Perkins, 1975). Thus, that the students have been fluent speakers and readers of English for most of their lives and that perception among texts is similar allows comparison of these groups. The mean age was 20 1. Fifteen subjects considered themselves good at reading upside down.

Table 1
ParagraphMean Standard Deviation
Words128.1 3.98
Characters586.9 42.14
Number of Lines 12.32.9
Percent Passive 0.0%0.00%
Flesch Reading Ease 80.113.72
Flesch-Kincaid Grade Level 3.92.06
Coleman-Liau Grade Level 52.725.16
Bormuth Grade Level 11.00.36

Analysis. Data was analyzed according to percentage changes in reading time, thus reducing any inter-subject differences in absolute reading ability (Figure 1). Mean times for each angle across all subjects were calculated and graphed (Figure 2). Paragraph times were extrapolated from the data and analyzed (Table 2, Figure 3).

Figure 1


Figure 2

Figure 3

Table 2
angle12 3456 78
01.001.00 1.001.001.00 1.001.001.00
451.010.94 1.210.860.92 1.021.310.96
901.441.47 1.341.311.20 1.141.631.40
1351.701.59 2.101.601.82 1.221.981.91
1801.791.43 1.661.721.27 1.701.552.17
2252.821.98 1.861.541.52 1.491.841.56
2701.301.66 1.531.251.06 1.071.341.15
3150.950.99 1.191.090.99 0.891.090.91

RESULTS

Eleven out of the 22 subjects displayed trends peaking on both sides of 180°. Mean troughs were significant among subjects at p0.10 from 135° to 180° and p<0.01 from 180° to 225° for a paired t-Test for two sample means for a hypothesized difference of 0.0 and alpha of 0.5. Four of the eight paragraphs showed peaks at 180° rather than troughs. Most subjects paused before beginning each reading to collect themselves for approximately one second. Paragraphs 1,4,6, and 8 showed peaks around 180 rather than troughs.

DISCUSSION

The time to perform implicit mental rotation of shapes increases in proportion to increases in angle between standard and comparison (Shepard and Metzler, 1971). I conjecture that as letters are perceived individually in rotated form(Larsen, 1984), they may fall under the same rules of perception as shapes. According to Kolers and Perkins (1975), the data and Aulhorn's work creates a problem in the theorizing that text is only rotated on the three coordinate axes by showing that a 180° is easier to read than lesser rotations. This reasoning is faulty as the rotations in Shepard and Metzler's work was merely a comparison rotation. Reading rotated text is a qualitatively different task from recognizing it. As Kolers and Perkins (1975, p.260) point out, "letters like 'p, b' or 'n, u,' are usually read according to the prevailing transformation, but may be misread. Their misreading reveals an erroneous application of the feedback mechanism described, where a different transformation was tried and the results, because of an ambiguous shape, nevertheless passed the recognition test. This phenomenon argues, as we have explained above, that testing and feedback are recurrent procedures active throughout the reading, rather than just once at the beginning." Furthermore, Shepard (1971) found that adults rotated shapes 60° per second. A person may look at a page of text and instantly know the orientation of the text at any angle, as observed in this study. Of course this may be a function of mental bulk; his shapes were ten cubes connected together, significantly bulkier than letters.

"Good readers were significantly faster readers in the upright condition, perhaps as a result of a shift from a whole-word reading method in the upright position to a strategy of single letter analysis in the inverted position" as supported by the recordings of eye movements. The decrease of good reader's time with rotated text is due to a change in strategy. That no significant change in reading time occurred with rotation for poor readers indicates no change in strategy. Both readers made fewer errors in the inverted cases (Larsen, 1984). No rectification of the data can be made at this time. Further experiments and analyses must be carried out.

Design problems. The study was fraught with design problems. The literature on the subject was much more rich than anticipated, and thus, the experiment was not well-tuned toward complementing the literature. Subjects were not a homogenous group. Paragraphs contained grammatical errors, proper names, repetitions of phrases, and were all written in different styles. There was no accounting for the practice effect of being presented angles in a step-wise fashion. There were no written instructions for subjects, and thus each received slightly different instructions. Lastly, the recorded times were at the experimenter's discretion and subject to error.

Onward. More accurate, more conclusive, more extensive studies are necessary to conclude this issue. Future experiments may involve case studies of dyslexics and how malfunctioning moduli give insight into the mechanisms of healthy individuals.


REFERENCES

Horton, Keith D. (1985) The Role of Semantic Information in Reading Spatially Transformed Text. Cognitive Psychology, 17(1), 66-88.

Kolers, P.A., & Perkins, D.N. (1975) Spatial and Ordinal Components of Form Perception and Literacy. Cognitive Psychology, 7(2), 228-267.

Larsen, Steen. Parlenvi, Paul. (1984) Patterns of Inverted Reading and Subgroups in Dyslexia. Annals of Dyslexia, 34, 195-203.

Shepard, R.N., & Metzler, J. (1971) Mental Rotation of three-dimensional objects. Science, 171, 701-703.

[1] http://www.rjgeib.com/thoughts/brotherhood/brotherhood.html

[2] http://www.rjgeib.com/thoughts/steinbeck/steinbeck.html

[3] http://www.rjgeib.com/thoughts/abraham/abraham.html

[4] http://www.rjgeib.com/thoughts/cultural/cultural.html

[5] http://www.rjgeib.com/thoughts/fate/fate.html

[6] http://www.rjgeib.com/thoughts/information/information.html

[7] http://www.rjgeib.com/thoughts/dp-society/dp-society.html

[8] http://www.rjgeib.com/thoughts/truth/truth.html

APPENDIX

Table A-1
Paragraph1 23 45 67 8
Words122.0 129.0131.0 128.0132.0 131.0130.0 122.0
Characters524.0 531.0560.0 614.0624.0 626.0601.0 615.0
Number of Lines 10.010.0 16.010.0 12.017.0 13.010.0
Percent Passive 0.0%0.0% 0.0%0.0% 0.0%0.0% 0.0%0.0%
Flesch Reading Ease 73.281.3 101.364.1 72.590.4 93.265.0
Flesch-Kincaid Grade Level 4.93.8 0.96.8 4.42.8 1.55.9
Coleman-Liau Grade Level 30.746.6 43.798.4 83.231.7 33.553.9
Bormuth Grade Level 10.811.1 11.111.4 11.111.1 10.211.1

1

I often ask people what is their definition of a "writer." They

usually tell me: "Somebody who has published books or articles"

or "A person who makes their living by writing." I disagree. I think

somebody who writes everyday and looks at the workaday world

through the prism of literature and words is a writer. Period. If a

person everyday honestly strives to get at the heart of a matter

and glean the truth out of the complexity of this world and we

human beings through the written word, than that person is a

writer, in my opinion. It is a matter of attitude and a way of life,

not how you pay your bills or whether you have talent or not.


2

The basic rule given us was simple and heartbreaking. A story to be

effective had to convey something from the writer to the reader, and the

power of its offering was the measure of its excellence. Outside of that,

there were no rules. A story could be about anything and could use any

means and any technique at all - so long as it was effective. As a subhead

to this rule, it seemed to be necessary for the writer to know what he

wanted to say, in short, what he was talking about. As an exercise we were

to try reducing the meat of our story to one sentence, for only then could

we know it well enough to enlarge it to three- or six- or ten-thousand

words.


3

So Abram rose, and clave the wood, and went,

And took the fire with him, and a knife.

And as they sojourned, both of them together,

Isaac the first-born spake, and said, My Father,

Behold the preparations, fire and iron,

But where the lamb for this burnt-offering?

Then Abram bound the youth with belts and straps,

And builded parapets the trenches there,

And stretched forth the knife to slay his son.

When lo! an angel called him out of heaven,

Saying, Lay not thy hand upon the lad,

Neither do anything to him. Behold,

A ram, caught in a thicket by its horns;

Offer the Ram of Pride instead of him.

But the old man would not so, but slew his son,

And half the seed of Europe, one by one.


4

"This books is intended for the use of those who can still be charmed by

books and who have an irreducible interest in the depiction of love. Books

about love inform and elevate the fantasy of their readers and actually

become part of their eros while teaching them about it... They are living

expressions of profound experiences... In itself and immediately this

transports us out of our dreary times. I hope that by this book I may touch

at least a few potential friends who can love literature in spite of the false

doctors [radicals] who try to cure them of it... This book bears witness to a

confrontation between the two greatest philosophical teachings about eros,

another chapter in the quarrel between the Ancients and the Moderns."


5

Tolstoy thought of it in just this way, but he did not spell it out

so clearly. He denied that history was set in motion by Napoleon

or any other ruler or general, but he did not develop his idea to

its logical conclusion. No single man makes history. History

cannot be seen, just as one cannot see grass growing. Wars and

revolutions, kings and Robespierres, are history's organic

agents, its yeast. But revolutions are made by fanatical men of

action with one-track minds, geniuses in their ability to confine

themselves to a limited field. They overturn the old order in a few

hours or days, the whole upheaval takes a few weeks or at most

years, but the fanatical spirit that inspired the upheavals is

worshipped for decades thereafter, for centuries.


6

The Eagle soars in the summit of Heaven,

The Hunter with his dogs pursues his circuit.

O perpetual revolution of configured stars,

O perpetual recurrence of determined seasons,

O world of spring and autumn, birth and dying!

The endless cycle of idea and action,

Endless invention, endless experiment,

Brings knowledge of motion, but not of stillness;

Knowledge of speech, but not of silence;

Knowledge of words, and ignorance of the Word.

All our knowledge brings us nearer to death,

But nearness to death no nearer to God.

Where is the Life we have lost in living?

Where is the wisdom we have lost in knowledge?

Where is the knowledge we have lost in information?

The cycles of heaven in twenty centuries

Brings us farther from God and nearer to the Dust.




7

"We don't read and write poetry because it's cute. We

read and write poetry because we are members of the human

race. And the human race is filled with passion. And

medicine, law, business, engineering, these are noble

pursuits and necessary to sustain life. But poetry, beauty,

romance, love, these are what we stay alive for. To quote

from Whitman, 'O me! O life!... of the questions of these

recurring; of the endless trains of the faithless--of cities

filled with the foolish; what good amid these, O me, O life?'

Answer. That you are here--that life exists, and identity;

that the powerful play goes on and you may contribute a

verse. That the powerful play goes on and you may

contribute a verse. What will your verse be?"


8

"The essence of civilization is the orderly quest for truth, the rational

perception of reality and all its facets, and the adaptation of man's

behaviour to its laws. So long as we follow the path of reason we shall not

move far from the lighted circle of civilization. Its enemies invariably lie

among those who, for whatever motive, deny, distort, minimize, exaggerate

or poison the truth, and who falsify the processes of reason. At all times

civilization has its enemies, though they are constantly changing their

guise and their weapons. The great defensive art is to detect and unmask

them before the damage they inflict becomes fatal. 'Hell.' wrote Thomas

Hobbes, 'is truth seen too late.' Survival is falsehood detected in time.