A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Experimental Psychology
This content is subject to copyright. Terms and conditions apply.
Journal
oj
Experimental
Psychology
1973,
Vol. 100,
No.
2,
259-262
ANAGRAM
SOLUTION
AS A
FUNCTION
OF
DIGRAM
VERSATILITY1
ROBERT
L.
SOLSO,2
GENE
K.
TOPPER,
AND
WILLIAM
H.
MACEY
Loyola University, Chicago
Bigram
versatility
(BV),
as
measured
by the
number
of
different
words
in
which
a
bigram
appeared,
was
established.
Anagram solution
as a
function
of BV and
bigrain
frequency
was
tested.
An
inverse
relationship
between
BV and
ana-
gram solution
time
was
observed.
The
results
were
discussed
within
the
con-
text
of a
reduced
search area
for
anagram solving.
The
solution
time
required
for un-
scrambling
anagrams
has
been
actively
studied.
Several
significant
variables
have
been
isolated
as
contributing
to
anagram
solution
time
(Kendler,
1961;
Kencller
&
Kendler,
1962).
Mayzner
and
Tresselt
(1958,
1959, 1962, 1963, 1966)
identified
anagram solution
to be a
function
of
(«)
word frequency
of the
solved word,
(b)
word
length,
(c)
number
of
letter
moves
from
anagram
to
solution word,
and
(d)
multiple
solutions
of
anagrams.
Mayzner
and
Tresselt
(1959, 1966) also reported
that
the
frequency
of a
bigram
in the
solution
word
may
affect solution time.
It is
plausible
that
anagrams containing
Q
and
U
or T and H,
which
are
frequently
combined
in a
bigram
in the
English lan-
guage, would provide
61
with
a
potent
cue
in
solving
an
anagram.
The
significance
of
bigram frequency
(BF)
and
anagram solution
has
also
been
studied
by
Dominowski
and
Duncan
(1964),
who
reported
data
seemingly
in-
consistent
with
that
of
Alayzner
and
Tres-
selt's
(1959, 1966) data.
The
difficulty
in
obtaining consistent
findings may be
due
to the
presence
of an
unisolated
variable.
Dominowski (1967) suggested that
Ss
used
bigrams
to
generate
words
as
pos-
sible
solutions
and the
higher
the
proba-
bility
of
bigram appearance
in the
solution
word
the
higher
the
likelihood
of
rapid
solution.
Dominowski's
logic
is
that
S
selects
one
letter
from
the
anagram
as
the
1
The
authors
express
their
appreciation
to
Carl
P.
Duncan
and
Roger Dominowski,
who
made
helpful
suggestions
on the
manuscript.
2
Requests
for
reprints should
be
sent
to
Robert
L.
Solso,
Department
of
Psychology, Loyola Uni-
versity
of
Chicago,
6525
Xorth
Sheridan Road,
Chicago,
Illinois
60026.
first
letter
of a
bigram
and
then matches
it
with
the
other
letters
in the
anagram
on
the
basis
of the
frequency
of the
bigram
(e.g.,
TH)
;
then presumably
the
search
for
the
solution
word
is
restricted
to
words
containing
TH. By
reducing
the
number
of
alternatives
that
S
must
test,
the
solu-
tion
is
expedited. However, Dominowski
noted
that
in
addition
to BF the
number
of
different
words
in
which
a
given
bigram
may
occur could also
affect
anagram solu-
tion
time.
We
propose
that
in
addition
to BF
bigrain
versatility (BV), defined
as the
number
of
different
words
in
which
a
given
bigram
may
occur,
is
also
an
important
variable.
If 5
generates bigrams
and
searches
for
words
that
may
contain
the
bigrams,
then
the
number
of
alternatives
that
he
must
consider,
as
determined
by
BV,
should
affect
his
solution time.
It is
noted
that
BV and BF may be
high cor-
relates
of
each other
but
that
they rep-
resent
significantly
different
dimensions.
To
illustrate
the
difference between
BV
and BF,
consider
the
bigram
OF.
This
has a
high
frequency
in the
English lan-
guage
but its
frequency
is
largely derived
from
the
word "of."
On the
other hand,
BV
tends
to be
relatively low.
The
effect
of BV in
anagram solving
may
be
related
to the
6"s
memory search
in
an
effort
to find a
match
between
bi-
grams (which
may
have
been formed
on
the
basis
of BF) and
words
which
contain
the
bigrams.
Thus,
the
higher frequency
bigrams with lower versatilities should
generate
fewer
potential
solution
words
(less
interference) than their
high-fre-
quency,
high-versatility counterparts.
The
present
study
was
designed
to
study
the
259
