Edward L. Thorndike's Law of Effect is a fundamental principle in psychology that explains how consequences influence learning. While it has faced skepticism, its empirical investigation remains crucial for understanding behavior modification. This article delves into Thorndike's own empirical studies that aimed to provide a "crucial test" of whether the after-effects of a connection truly strengthen or weaken it, making it easier for students to understand Thorndike's Law of Effect: Empirical Study.
The Law of Effect: An Enduring Hypothesis
Thorndike's Law of Effect posits that responses followed by satisfying consequences are more likely to be repeated, while those followed by annoying consequences are less likely. This empirical hypothesis was initially developed to explain learning in animals, especially when simple frequency of an action couldn't account for changes in behavior. Over time, it was extended to human learning, encompassing the potency of wants, interests, purposes, and desires.
Historically, the Law of Effect suffered from disfavor among psychologists and neglect from physiologists, often deemed "troublesome to think with" compared to more mechanical forces like frequency and intensity. Philosophers and educational theorists also found it challenging, preferring teleological or logical explanations such as purposes or relief from tensions. However, recent experiments have underscored its importance, showing that mere frequency alone isn't always enough for learning, making a thorough investigation of the law more relevant than ever.
Experiment A: Estimating Lengths with "Right" or "Wrong" Feedback
Thorndike designed Experiment A as a direct test of how "satisfyingness" and "annoyingness"—specifically, the spoken words "right" and "wrong" from the experimenter—affected learning. The study involved participants estimating the lengths of paper strips.
Experimental Setup and Procedure for Length Estimation
The setup for Experiment A was meticulous:
- Materials: A series of 50 paper strips, with two of every unit length between 3 and 27 cm, were used. All strips were identical except for their length.
- Standard Reference: Participants had a fixed 10 cm strip, known to be 10 cm, for reference.
- Presentation: Strips were presented one at a time on a fixed background, in a random order, with no prior knowledge of the series constitution.
The experiment proceeded in distinct phases:
- Initial Judgments: 50 judgments were made with no aid beyond the 10 cm standard. No feedback was given.
- Training Phase: The core of the experiment. After each estimate, the strip was removed and turned over behind a screen. Approximately 2 seconds after the participant's announcement (and 1.5 seconds after the strip's removal), the experimenter stated "right" or "wrong" based on the accuracy of the estimate. No information about the amount or direction of error was provided. This phase typically involved seven such training periods over three to four days.
- Final Test: The series was presented again with no aid, similar to the initial judgments, to assess retention of learning.
Experiment A Results: Impact of Consequence
The results of Experiment A were striking, mirroring animal learning experiments where food or freedom acts as reward. The "satisfying connections" (correct estimates) were significantly strengthened, leading to a rapid rise in successful judgments. Participants showed a substantial reduction in total deviation from true lengths. For instance, across five subjects, the sum of deviations decreased from an average of 64 before training to 14 after, representing a 64% reduction in error. This was in stark contrast to control experiments where mere repetition without "right" or "wrong" consequences yielded little to no improvement.
Both constant errors (tendency to consistently overestimate or underestimate) and variable errors (inconsistency in judgments) were reduced, demonstrating a comprehensive improvement in accuracy. The study by Abelson, a control experiment, further solidified these findings, showing an average reduction of 50 units in deviation with feedback, compared to only 6 units with mere repetition.
Experiment B: Blindfolded Line Drawing and "Right"/"Wrong" Feedback
Experiment B offered another empirical test of the Law of Effect, focusing on motor skills and perception under different conditions.
Experimental Setup and Procedure for Line Drawing
This experiment involved participants drawing lines of specific lengths while blindfolded:
- Task: Draw 3-, 4-, 5-, and 6-inch lines.
- Setup: Participants were blindfolded and seated opposite the experimenter, with a drawing board and cross-section paper. A strip of veneer on the left edge served as a fixed starting point for all lines.
- Drawing Method: Participants were instructed to draw lines of a given length with one continuous, quick movement, starting from the veneer strip.
- Feedback: "Right" was announced if the line was within 1/8 inch of the target length for 3-inch lines, or 1/4 inch for other lengths. Otherwise, "wrong" was announced. Feedback came approximately one second after drawing each line.
- Training: Participants were trained at a single sitting, drawing various lengths and waiting for feedback before drawing the next line.
Experiment B Results: Learning Under Blindfolded Conditions
Similar to Experiment A, Experiment B demonstrated a clear influence of the "right" or "wrong" feedback. Every participant showed improvement during training. The median percentage of "right" responses increased from 13% in the early test to 54% in the seventh training period. Even in the late test, without guidance, the median remained at 26%, double the initial performance. The average gain in "right" responses was 12 percentage points.
More detailed analysis of the average errors for each line length confirmed this. A significant reduction in error was observed for each length and in 18 out of 24 subjects, with an average reduction nearly eight times its probable error. A control experiment with 6 subjects, who drew lines without any "right" or "wrong" announcements, showed only minimal improvement, further proving that the effects of a connection directly alter its strength, rather than mere repetition.
Direct vs. Indirect Influence of Effect
A critical question arising from these empirical studies is whether the after-effects of "right" or "wrong" influence connections directly, or indirectly by prompting internal mental repetitions. Thorndike considered both possibilities.
In Experiment B, where actual repetition of a successful movement could not occur (as the task was to draw new lines), the participant's most likely internal action would be to rehearse the "inner feel" or cues that led to a successful drawing after a "right" announcement, and to abandon or neglect them after a "wrong." While such deliberate retention or recall likely plays a part, Thorndike suggested a more fundamental, general, and direct favoring of successful connections.
His reasons for this direct influence include:
- Lack of Evidence for Deliberate Recall Superiority: There's no clear evidence that conscious effort to retain cues improves learning faster than simply performing the action.
- Animal Learning Analogy: It's unlikely that animals like cats, dogs, or rats engage in complex deliberate reviews or rehearsals during their learning processes.
Thorndike proposed that this "more fundamental, general, and direct favoring" could be as simple as a longer duration of the connection when "right" is heard, akin to the strengthening influence of more repetitions of equal length.
Flashcards
Tap to flip · Swipe to navigate
Characteristics of Learning in Thorndike's Studies
Two notable features of the blindfolded line-drawing task stood out:
- Slow and Gradual Improvement: After an initial stage where gross errors were eliminated, the process of refinement was slow and gradual. For example, the gain in "right" responses between the second and seventh practice periods was only from an average of 42% to 54%.
- Immediate Guidance and Abrupt Failures: When "right" or "wrong" guidance was removed, there was an immediate drop in performance, suggesting a heavy reliance on external feedback. Conversely, in situations where success is immediately and easily identifiable (e.g., drawing a catenary if one knows what it is), learning can be very sudden and complete.
These characteristics highlight the nuanced interplay between feedback, internal cognitive processes, and the lasting strength of connections as described by the Law of Effect.
FAQ: Understanding Thorndike's Law of Effect
Students often have questions about Thorndike's foundational work. Here are some common inquiries:
What is the core principle of Thorndike's Law of Effect?
The core principle is that responses that produce a satisfying or pleasant effect in a particular situation become more likely to occur again in that situation, while responses that produce an annoying or unpleasant effect become less likely. This creates a direct link between behavior and its consequences.
How did Thorndike empirically test the Law of Effect?
Thorndike conducted experiments like estimating lengths of lines (Experiment A) and drawing lines blindfolded (Experiment B). In these studies, participants received immediate "right" or "wrong" feedback for their actions. He then compared their learning progress to control groups who received no such specific feedback, demonstrating that the consequences significantly influenced learning and accuracy.
Why was the Law of Effect initially met with disfavor?
It was disfavored by some psychologists and neglected by physiologists because it was considered less "mechanical" and harder to conceptualize than simpler forces like frequency or intensity of stimuli. Philosophers and educational theorists also preferred explanations based on purpose, tension relief, or other teleological factors over a purely empirical, consequence-based mechanism for learning.
Did Thorndike believe the Law of Effect acted directly or indirectly on connections?
While he acknowledged that internal rehearsal of successful cues (an indirect effect) likely plays a part, Thorndike argued for a more fundamental, general, and direct favoring of successful connections. He suggested this direct influence might involve a longer duration of the connection itself when a positive consequence occurs, akin to the strengthening effect of increased repetition.
What do Thorndike's studies tell us about mere repetition versus consequences in learning?
Thorndike's experiments clearly demonstrated that mere repetition, without specific feedback or consequences, leads to very little or no improvement in performance. The significant gains in accuracy and reduction in errors observed in both Experiment A and Experiment B were directly attributable to the "right" or "wrong" feedback, underscoring the vital role of consequences in strengthening learning connections.