ArticlePDF Available

The odd effect and multiple meanings in English of the words "odd" and "even".

Authors:

Abstract

Under a variety of conditions, people take longer to make judgments about odd than about even digits and digit names. In English the words "odd" and "even" have multiple meanings. Perhaps the multiple meanings of these words are responsible for the slowing of responses to odd stimuli. This hypothesis was tested using participants who spoke no English and in whose native language, Polish, the words for the mathematical concepts of odd and even do not have multiple meanings.
Psychological Reports, 2010, 107, 2, 443-446. © Psychological Reports 2010
DOI 10.2466/04.10.11.22.PR0.107.5.443-446 ISSN 0033-2941
THE ODD EFFECT AND MULTIPLE MEANINGS IN
ENGLISH OF THE WORDS “ODD” AND “EVEN”1
TERENCE M. HINES
Psychology Department
Pace University
Summary.—Under a variety of conditions, people take longer to make judg-
ments about odd than about even digits and digit names. In English the words
“odd” and “even” have multiple meanings. Perhaps the multiple meanings of these
words are responsible for the slowing of responses to odd stimuli. This hypothesis
was tested using participants who spoke no English and in whose native language,
Polish, the words for the mathematical concepts of odd and even do not have mul-
tiple meanings.
            -
ditions, it takes people longer to make judgments about odd than about

           

“odd” is a linguistically marked concept while “even” is unmarked and
noted evidence that marked adjectives require greater processing time.
The unmarked adjective in a pair has a lower word frequency and can take
-
  -
tive pair before they learn the marked member (E. Clark, 1973; H. Clark,
1973). Hines reanalyzed data from two early studies (Clapp, 1924; Knight
& Behrens, 1928) of the ease with which children learned arithmetic prob-
lems. Both studies showed that children take longer to learn sums of two
    
100 single-digit addition problems. The average rank for the odd plus odd
problems was 79; for the even plus even problems it was 57. Knight and
Behrens’ rankings of the same problems were 83 for the odd and 68 for
the even. Conceptually, a marked adjective may be thought of as noting
the absence of the characteristic of the unmarked adjective, but not the re-
verse. Thus, dead is a marked adjective because death is the absence of life
and one can say undead but not unalive.
         
    
meanings. One is mathematical and indicates that an odd number is not
1Address correspondence to Terence Hines, Psychology Department, Pace University, Pleas-
antville, NY 10570-2799 or e-mail (thines@pace.edu).
T. M. HINES
444
evenly divisible by two. The other denotes something strange or unusual.
The word “even” also has a mathematical meaning and other meanings
(i.e., smooth). Perhaps something about the multiple meanings of these
         
about odd and even stimuli.
       
which the words denoting the mathematical meaning of “odd” and “even”
do not carry other meanings. Polish is such a language. In Polish, the word
for the mathematical concept odd is “neiparzysty” and has no connotation
of strangeness or other meanings as does “odd.” The Polish word mean-
ing strange or unusual is “dziwny.” The Polish word for the mathemati-
cal concept even is “parzysty” (Stanislawski, 1964; Stanislawski & Szercha,
1969). If a slower response to odd digits is associated with the multiple
meanings of odd and even in English, this slowing should not be present

tested this hypothesis. Hines, Herman-Jeglinska, Bednarek, and Grabow-

speaking participants, but many of these did speak English. However, that

-

include an English-speaking group in the present study. The goal of the

-
-
ing group, further research would be required to specify what it is about


Participants
Participants were 24 University of Warsaw undergraduates, 19 wom-
en and 5 men, whose mean age was 20.5 yr. (SD
made over 20% errors and their data were not used. Participants were na-
tive Polish speakers who spoke no English. They were paid the equivalent
in Polish currency of U.S. $3. for their participation.
Apparatus and Procedure
Testing was conducted by a native Polish-speaking psychology ma-
jor who read the instructions and answered any questions from partici-
pants. Testing took place at the Psychology Department of the University
of Warsaw. Participants sat at a self-selected distance in front of a monitor
controlled by an IBM P2, Model 50 computer, which controlled stimulus
presentation and response recording. Stimuli were of pairs of digits se-
MEANINGS OF ODD AND EVEN 445
lected from 2 through 9, separated by four blank spaces. Digit pairs were
presented centered on the computer screen. No identical digit pairs were
presented. There were 188 trials of which 47 consisted of two even and 47
of two odd digits. On the remaining 94 trials, one digit was odd and one

-
perimental trials.
Participants were instructed to press the A” key on the computer

        
was odd and one even. The crucial comparison in this study, between re-
sponses when the two stimuli were both odd or both even, was made by
the same hand. Participants were asked to respond as quickly and accu-
rately as possible.
When a correct response was given, feedback in the form of that trial’s
reaction appeared on the screen. If an error was made the word “ERROR”
appeared. If the reaction time was greater than 1,500 msec. “TOO SLOW”
appeared. This feedback was in English as no Polish character set was
    
    -
moved from the screen and the feedback appeared, remaining visible for
   
pair appeared.

 
both digits were odd than on even/even trials. Reaction time was slower,
1,017 msec. (SDSD-
 t18 p  d    -
ly more errors on odd/odd trials, 11.3% (5.4) versus 2.2% (1.4; t18
p  d      
was 973 msec. (SDSD
The analysis shows that odd digits were responded to more slowly
even in a language in which the words for the mathematical concepts odd
and even          -
        
  -
  -
fect. Neiparzysty is marked in Polish. In fact, it literally translates as uneven
or “not even.” Just as it is improper to say unodd in English, one cannot say
nei neiparzysty in Polish.
       -
plete. It does not address the issue of what aspects of the internal repre-
T. M. HINES
446
sentations of marked and unmarked concepts are important in the slow-
ing of responses to marked concepts. Perhaps the internal representations
          
memory than marked representations. This is an issue for future research.
REFERENCES
The number combinations: their relative diculty and the frequency of their
appearance in text-books. Bureau of Educational Research Bulletin No. 1. Madison,
WI: Univer. of Wisconsin Press.

language. In T. Moore (Ed.), Cognitive development and the acquisition of language.
New York: Academic Press. Pp. 65-110.
      Cognitive
development and the acquisition of language. New York: Academic Press. Pp. 27-63.

digits. Memory & Cognition, 18, 40-46.
&  -
ences in the processing of odd and even digits. Acta Neurobiologiae Experimentalis,
56, 263-266.
&   The learning of the 100 addition and 100 subtraction
combinations. New York: Longmans, Green.
The great English-Polish dictionary. Warsaw, Poland: Philip Wil-
son.
&Wielki slownik Polsko Angielski. Warsaw, Poland:
Philip Wilson.
Accepted August 19, 2010.
... Participants took longer to respond to odd numbers than to even numbers. The size of this odd effect depends on the format (Arabic digits vs. number words) and the type of the task used (Hines, 1990(Hines, , 2010. Next, when the hand of the response is considered, faster right-hand responses to even than odd numbers and faster left-hand responses to odd than even numbers have been observed (Berch et al., 1999;Nuerk et al., 2004;Reynvoet & Brysbaert, 1999;Willmes & Iversen, 1995). ...
... The SNARC and MARC effects were also found in an applied context in a card game scenario (Schroeder & Pfister, 2015). There are important cultural and individual differences that modulate the direction and strength of the SNARC (Bull et al., 2013;Cipora et al., 2016;Fischer, 2008;Shaki et al., 2009) or MARC effect (Hines, 2010;Huber et al., 2015). Interestingly, the SNARC and MARC effects do not correlate with each other, which was observed in the lab (Huber et al., 2015;Krajcsi et al., 2018) and in the online study (Cipora et al., 2019). ...
... Taken together, our results, the studies cited above and the finding that the SNARC and MARC effects do not correlate with each other (Cipora et al., 2019;Huber et al., 2015;Krajcsi et al., 2018) support the conclusion that they are driven by different mechanisms. Cultural factors can also contribute to the representation of parity (Heubner et al., 2018;Hines, 2010;Nishiyama, 2006). For example, the English words even and odd carry multiple meanings sometimes associated with positive and negative evaluations, respectively. ...
Article
Full-text available
In two experiments we tested the contribution of linguistic markedness and affective evaluation (i.e., body specificity) to the representation of abstract numerical concepts, such as parity. To this end, we employed speeded parity judgments of digits (Exp 1) or number words (Exp 2) in a go/no-go task. Fifty right-handed participants completed four blocks of trials in each experiment. In two blocks, they responded to even numbers (2, 4, 6, or 8) and in the other two blocks they responded to odd numbers (1, 3, 7, or 9). In each pair of blocks, they responded once with their right hand and once with their left hand. Results revealed faster right-hand responses to even than to odd digits (Exp 1), and faster left-hand response to odd than to even number words (Exp 2). In addition, in both experiments, we found faster responses to small-odd than large-odd digits and number words. The results support the conclusion that the affective evaluation of parity and linguistic markedness makes independent contributions to the representation of parity.
... Literature has shown that humans perceive and think about odd and even numbers differently; our brains take longer to react to and process odd numbers [13]. It is hypothesized that this is due to the linguistic markedness and multiple definitions of the word "odd", potentially causing people to view even numbers as the norm [14]. ...
Article
Full-text available
Background: Actinic keratosis (AK) is a pre-cancerous skin condition caused by sun exposure. Number bias, a phenomenon that occurs when meaning other than numerical value is associated with numbers, may influence the reporting of AK removal. The present study aims to determine if number bias is affecting healthcare providers' documentation of patient-provider encounters. Methods: A single-center retrospective chart review of 1415 patients' charts was conducted at the University of Cincinnati Medical Center. To determine if there was a significant difference between even and odd-numbered AK removals reported, an exact binomial test was used. The frequency of removals per encounter was fitted to a zero-truncated negative binomial distribution to predict the number of removals expected. All data were analyzed with RStudio. Results: There were 741 odd and 549 even encounters. Odd removals were reported at a significantly greater frequency than even p < 0.001. Age may be contributing to the observed number bias (p < 0.001). One, two, and eight were reportedly removed more frequently, while nine, 13, and 14 were reportedly removed less frequently than expected, respectively. Conclusion: Number bias may be affecting clinicians' documentation of AK removal and should be investigated in other clinical settings.
Article
Full-text available
12 males and 8 females made odd vs even classification judgments on numbers presented visually in 3 different formats: digits, number words, and dot patterns. Males showed longer decision times for odd numbers only when the numbers were presented in the dot pattern format. Females showed this effect only when the stimuli were presented in the word format. These results suggest that a differential speed of response to odd and even numbers is found most strongly when the stimuli are presented in a format which is processed more efficiently by the Ss. This finding may imply that the effect is being produced by some higher order cognitive process, based on higher order representations that do not necessarily involve linguistic or verbal coding. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Eight experiments are reported that first establish and then explicate a serendipitous finding that judgments about whether digits are odd or even take longer for odd than for even digits. The slowing of judgments about odd digits is more pronounced when digit pairs or triples are used, but is still weakly present when a single digit must be classified. A similar effect is seen when judgments of nouns are based on whether the nouns are the names of living or dead objects. Nouns that name dead objects are judged more slowly than ones that name living objects. The concept "alive" is linguistically marked. Past research has shown that unmarked concepts are processed more rapidly than marked ones. The similarity in the pattern of results when digits and words are judged is used to argue that the slower judgments about odd digits are due to the fact that "odd" is a linguistically marked and "even" a linguistically unmarked concept.
Chapter
This chapter focuses on of the knowledge that one has to have about a word to use it appropriately. From the developmental point of view, what the child knows about the meaning of a word needs to be found in addition to the way in which this knowledge changes during the language acquisition process. The semantic feature hypothesis assumed that the meanings of words are made up of features or components of meaning and proposed that children learn word meanings gradually by adding more features to their lexical entries. The general predictions made by this theory have been shown to be remarkably consistent with data from several different sources in the literature on children's language. The theory contains a number of lacunae that future work will have to fill. For example, there is no account of the internal structure or lack of it in the child's earliest lexical entries. To study language acquisition properly, semantics cannot be ignored, for it is essential to know what the children means by what they says, and to know how they understand what they hear.
Article
"The problem is in question form: How much practice is needed by typical second-grade children to learn the 100 addition and the 100 subtraction combinations? A minor problem is: How much practice is needed by second-grade children to keep from forgetting these 200 learned basic facts?" The learning of each combination by each child was directed, and also recorded in terms of accuracy, of rate, of maintenance, and of errors in the connections formed. Upon evidence resulting from intensive study of 25 and of 15 children's responses yielding over 100,000 and 60,000 specific data for addition and subtraction respectively, tables are presented revealing "general difficulty rank of each of the combinations, the number of practices needed by the 50th and by the 80th percentile pupil to master them, the average time of practices for permanence, and the average errors (number and instances) made in learning. Intercorrelations reveal the inner consistency of these data. These Knight-Behrens difficulty rankings of addition and of subtraction combinations (1) in comparison with those of Clapp, who based his rankings upon the study of errors (inherent difficulties plus unknown amounts of previous practices) of fifth-grade children, show a correlation of approximately + .66; (2) but in contrast with the Clapp rankings, the Knight-Behrens rankings square up with other known facts of human learning, in that they substantiate relationship: (a) between the difficulty of an addition combination and its reverse (explained to some extent by the hypothesis 'the size of the sum,' and by the table making more 'summarized' thinking possible), (b) between companion subtraction combinations, and (c) between related addition and subtraction facts." (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The number combinations: their relative difficulty and the frequency of their appearance in text-books
  • F Clapp
Clapp, F. (1924) The number combinations: their relative difficulty and the frequency of their appearance in text-books. Bureau of Educational Research Bulletin No. 1. Madison, WI: Univer. of Wisconsin Press.
Cognitive development and the acquisition of language
  • H Clark
Clark, H. (1973) Space, time, semantics, and the child. In T. Moore (Ed.), Cognitive development and the acquisition of language. New York: Academic Press. Pp. 27-63.
Wielki slownik Polsko Angielski
  • J Stanislawski
  • M Szercha
Stanislawski, J., & Szercha, M. (1969) Wielki slownik Polsko Angielski. Warsaw, Poland: Philip Wilson.