ERIC EJ941322: College-Level Instruction: Derived Relations and Programmed Instruction

2011·0.13 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2011, 44, 413–416 NUMBER2 (SUMMER2011)
COLLEGE-LEVEL INSTRUCTION: DERIVED RELATIONS AND
PROGRAMMED INSTRUCTION
DANIEL M. FIENUP
QUEENSCOLLEGEANDTHEGRADUATECENTEROFTHECITY
UNIVERSITYOFNEWYORK
JEFFERY HAMELIN AND KIMBERLY REYES-GIORDANO
THEGRADUATECENTEROFTHECITYUNIVERSITYOF
NEWYORK
AND
TERRY S. FALCOMATA
UNIVERSITYOFTEXASATAUSTINANDTHEMEADOWS
CENTERFORPREVENTINGEDUCATIONALRISK
Recent research has demonstrated the effectiveness of programmed instruction that integrates
derivedrelationstoteachcollege-levelacademicmaterial.Thismethodhasbeendemonstratedto
be effective and economical in the teaching of complex mathematics and biology concepts.
Althoughthisapproachmayhavepotentialapplicationswithotherdomainsofcollegelearning,
more studies are needed to evaluate important technological variables. Studies that employ
programmed instructionare discussed inrelation tofuturedirections for research.
Keywords: college students, derived relations, programmed instruction
_______________________________________________________________________________
Recent advances in college-level instruction and stimulus types might consist of class name
have included programming for derived rela- (i.e., mammal and reptile), skin type (i.e., furry
tions. A derived relation is an outcome that and scaly), birth method (i.e., live and egg), and
results from directly teaching the interconnec- thermophysiology (i.e., warm-blooded and cold-
tedness of stimuli. The most important type of blooded). Direct teaching of class names with
derived relation involves relating two or more each stimulus type would likely result in the
stimuli that have never been directly paired but emergence of relations between the stimuli
share an association with a common stimulus without directly teaching those relations. These
(Critchfield & Fienup, 2008). Programming applications are economical in that individuals
for derived relations is accomplished by orga- demonstrate adaptive behavior in excess of that
nizing stimuli according to classes and types of which was directly taught (Fienup & Critch-
stimuli (e.g., see Figure 1 of Fields et al., 2009) field, 2010; Stromer, MacKay, & Stoddard,
and determining what is likely to emerge, given 1992; Toussaint & Tiger, 2010).
therelationsdirectlytaught.Forexample,when Both relational frame theory (RFT, Hayes,
teaching about different classes in biology, Barnes-Holmes,&Roche,2001;Ninnessetal.,
classes might consist of mammal and reptile, 2009) and stimulus equivalence (Fields et al.,
2009) paradigms have accounted for derived
Address correspondence to Daniel M. Fienup, Depart- relations. From an application standpoint, both
mentofPsychology,QueensCollege,65-30KissenaBlvd., paradigms are inherently interested in classes of
Flushing, New York 11367 (e-mail: daniel.fienup@
physically different stimuli that are functionally
qc.cuny.edu).
doi:10.1901/jaba.2011.44-413 related. Instructional applications from both
413
414 DANIEL M. FIENUP et al.
paradigms involve teaching the minimal num- of48relations(i.e.,fourtimesasmanyrelations
ber of relations between members of a class so learned as taught). The authors also found that
that all relations between stimuli emerge in an students were able to generalize responding to
efficient manner. novel graphs.
Additional examples were provided by
College-Level Instructional Research on
Fienup and Critchfield (2010) and Critchfield
Derived Relations
and Fienup (2010) in teaching concepts of
Typically, derived relations research has
inferential statistics and hypothesis decision
involved instruction on how various represen-
making.Inthefirstlesson,studentsweretaught
tations of stimuli are positively related. The
two statistical categories: one that involved
majority of these applications have focused on
statistically significant stimuli and one that
statistical and mathematical concepts. For
involvestatisticallynonsignificantstimuli.Next,
example, Ninness et al. (2006, 2009) pro-
students were taught to match hypothesis- and
grammed for derived relations when teaching
results-basedstatementstostatementsaboutthe
college students trigonometry and algebraic
null hypothesis (reject or fail to reject) and
concepts. The instructional package included
scientific hypothesis (consistent or inconsis-
verbal rules and match-to-sample training. The
tent). Last, students were taught to incorporate
stimuli involved standard equations (y 5 6 2
statistical information with hypotheses and
!2x24),factoredequations(y5!2[x+4]+
results to make accurate decisions about
6),andgraphicalrepresentations.First,students
hypotheses. In total, students were directly
were taught rules about how the stimuli were
taught 40 relations and demonstrated up to
related. Next, students received computerized
144 relations (i.e., 4.7 times as many relations
match-to-sample training of the respective
learned as taught). In addition to these
stimuli. Following instruction, students consis-
demonstrations of the application of stimulus
tentlydemonstratednovelformsofbehavior,or
equivalence to math instruction, similar appli-
derived relations, such as matching graphs with
cations have also been demonstrated with
standard equations. In addition, they general-
brain–behavior concepts (e.g., Fienup, Covey,
ized responding to novel types of functions.
& Critchfield, 2010) and disability categoriza-
Recently,Ninnessetal.(2009)incorporatedthe
tion (Walker, Rehfeldt, & Ninness, 2010).
construction of graphs. This approach allowed
students to complete complex reciprocal trigo-
Future Directions
nometric relations.
Fields et al. (2009) taught college students Several areas are in need of future research.
conceptsofstatisticalinteractions.Studentswere For example, a number of new content areas
taughtfouruniquecategoriesofinteractions(no could benefit from instruction that programs
interaction,divergent,synergistic,crossover)and for derived relations. Areas such as geography,
four different representations of the category the taxonomy of biological organisms, psychi-
(graph, a description of the interaction repre- atric disorder categorization, and language all
sented in the graph, name of interaction, and involvemultiplerepresentationsofconceptually
definition of interaction). Using only match-to- relatedstimuli.Someoftheseareashavealready
sample training, the experimenter taught the been investigated with young learners (e.g.,
students three relations. Following teaching, geography,LeBlanc,Miguel,Cummings,Gold-
students generalized responding to novel stimu- smith,&Carr,2003;foreignlanguage,Joyce&
lus relations, including matching graphs and Joyce, 1993) and could be extended to include
definitions. Instruction involving 12 relations more complex topics that are taught in college
(threerelationsbyfourclasses)resultedinatotal curricula.
BRIEF REVIEW 415
A significant practical challenge involves the It is important to determine which of these
technology for promoting derived relations. inputs are necessary and sufficient when
Other areas within applied behavior analysis designing efficient instruction. It is not enough
have agreed-upon technology (e.g., functional to claim an instructional package is efficient by
analysis, Hanley, Iwata, & McCord, 2003), but illustrating that the stimulus relations output
there is no consensus in the literature on what exceeds the input. The actual teaching portions
experiences a college student requires to derive of these computerized programs are likely
relations reliably. The lack of consensus on how efficient. What has been given less attention is
to program for derived relations can be a the intensive nature of the testing batteries.
deterrent to new experimenters and instructors Measuring the emergence of derived relations is
whowanttoprogramforderivedrelations.Areas typically done across many trials (e.g., Fienup
of divergence in the literature include the et al., 2010, involved a minimum of 448 test
sequence of teaching, the sequence of testing trials). Programmed instruction is a package,
and training, the content of testing, the mastery and research is needed to determine how to
criterion,andthetypeandfrequencyoffeedback. measure the emergence of derived relations
This issue can be illustrated by comparing accurately while maintaining an efficient pro-
and contrasting research. Ninness et al. (2006) gram of instruction.
combinedmatch-to-sample trainingwithverbal College-level instruction that programs for
rules about how the stimuli were related, derived relations has shown promise for future
whereas Fields et al. (2009) used match-to- applications. This type of instruction, whether
sample training alone. Both used linear-series using an RFT or stimulus equivalence para-
training (e.g., ARB, BRC), but Ninness et al. digm, can be applied to a variety of academic
preceded and followed this training with mixed contents. However, more research is needed to
tests of all relations. Fields et al., on the other refine the technology of programming for
hand, interspersed additional tests before, derived relations in instructional settings, espe-
during, and after training to measure the cially if the goal is wider acceptance, use, and
gradual emergence of derived relations. application of derived relations.
The field has no agreed-upon approach for
even seemingly simple parameters such as type REFERENCES
and frequency of feedback. Fields et al. (2009)
Critchfield, T. S., & Fienup, D. M. (2008). Stimulus
providedvisualcorrectivefeedbackoneachtrial
equivalence. In S. F. Davis & W. F. Buskist (Eds.),
and then faded feedback across trials. These 21st century psychology (pp. 360–372). Thousand
Oaks,CA:Sage.
authors evaluated mastery based on blocks of
Critchfield, T. S., & Fienup, D. M. (2010). Using
trials, meaning that after the passage of a set
stimulus equivalence technology to teach about
number of trials, the computer program statistical inference in a group setting. Journal of
evaluated whether a criterion had been met Applied BehaviorAnalysis,43,437–462.
Fields, L., Travis, R., Yadlovker, E. D., Roy, D., Aguiar-
since the last evaluation. More recently, Fienup
Rocha, L., & Sturmey, P. (2009). Equivalence class
et al. (2010) incorporated a rolling mastery formation: A method for teaching statistical interac-
criterion. Participants were required to respond tions. Journal of Applied Behavior Analysis, 42,
575–593.
correctly on a certain number of consecutive
Fienup,D.M.,Covey,D.P.,&Critchfield,T.S.(2010).
trials. These authors also provided auditory
Teaching brain–behavior relations economically with
feedback and visual feedback that allowed stimulus equivalence technology. Journal of Applied
participants to view how many trials were on Behavior Analysis,43,19–33.
Fienup, D. M., & Critchfield, T. S. (2010). Efficiently
a test or how many consecutive trials on which
establishing concepts of inferential statistics and
they had responded correctly. hypothesis decision making through contextually
416 DANIEL M. FIENUP et al.
controlled equivalence classes. Journal of Applied Ninness,C.,Dixon,M.,Barnes-Holmes,D.,Rehfeldt,R.
BehaviorAnalysis,43,437–462. A., Rumph, R., McCuller, G., et al. (2009).
Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Constructing and deriving reciprocal trigonometric
Functional analysis of problem behavior: A review. relations: A functional analytic approach. Journal of
JournalofAppliedBehavior Analysis,36,147–185. AppliedBehaviorAnalysis,42,191–208.
Hayes, S. C., Barnes-Holmes, D., & Roche, B. (Eds.). Stromer, R., Mackay, H. A., & Stoddard, L. T. (1992).
(2001). Relational frame theory: A post-Skinnerian Classroom applications of stimulus equivalence
account of human language and cognition. New York: technology. Journal of Behavioral Education, 2,
KluwerAcademic/Plenum. 225–256.
Joyce, B. G., & Joyce, J. H. (1993). Using stimulus Toussaint, K. A., & Tiger, J. H. (2010). Teaching early
equivalenceprocedurestoteachrelationshipsbetween braille literacy skills within a stimulus equivalence
EnglishandSpanishwords.Education&Treatmentof paradigm to children with degenerative visual im-
Children,16,48–65. pairments. Journal of Applied Behavior Analysis, 43,
LeBlanc, L. A., Miguel, C. F., Cummings, A. R., 181–194.
Goldsmith, T. R., & Carr, J. E. (2003). The effects Walker, B. D., Rehfeldt, R. A., & Ninness, C. (2010).
of three stimulus-equivalence testing conditions on Using the stimulus equivalence paradigm to teach
emergent US geography relations of children diag- course material in an undergraduate rehabilitation
nosed with autism. Behavioral Interventions, 18, course. Journal of Applied Behavior Analysis, 43,
279–289. 615–633.
Ninness, C., Barnes-Holmes, D., Rumph, R., McCuller,
G., Ford, A. M., Payne, R., et al. (2006). Transfor- Received August 30,2010
mations of mathematical and stimulus functions. Final acceptance October21,2010
JournalofAppliedBehavior Analysis,39,299–321. Action Editor,Dorothea Lerman