Task Complexity and Goal Specificity

Goal specificity is central to goal-setting theory. I would expect task complexity to have a significant impact on the optimal degree of goal precision: the more complex an activity - hence the less clear the effort-performance link is - the more arbitrary and risky I assume to be the setting of very specific goals, given that the individual cannot directly link his inputs to the goal attainment. Is in those cases more effective to shift from setting goals as specific-result parameters to setting more generic targets in line with underlying interests?

Goal setting theory states that assigning difficult, specific goals on simple or routine tasks results in performance increases. However, today 's organizations are characterized by increasing complexity, therefore a critical issue in work settings is how to help people perform well when dealing with difficult and complicated tasks.
Goal orientation theory may help address this issue. Goal orientation refers to two types of super-ordinate goals people can hold during task performance: a learning goal orientation[i] and a performance goal orientation[ii]; while in the former the individual is not directly focused on the concrete final outcome of his performance, in the latter he is specifically concerned with the results achieved. Substantial research[iii] suggests that one goal orientation might not be superior in all circumstances, and that several factors including the cognitive load of the task - determined by its complexity and difficulty - might play a role in the efficacy of different goal orientations.
Steele-Johnson et al. claimed that a learning goal orientation is relatively more beneficial when cognitive load is high and vice versa. In their study, they examined the role of task complexity in goal orientation effects on performance, motivation, and satisfaction; their threefold hypothesis was that goal orientation and task complexity interact affecting performance, intrinsic motivation, and satisfaction with performance. Their study was based on a sample of 199 undergraduate students (79 M, 120 W) from a mid-western university and used a 2 (simple/complex task) X 2 (learning/ performance goal orientation) X 4 (trial blocks) design. All the hypotheses were tested using a 2 (task complexity) X 2 (goal orientation) X 4 (trial blocks) repeated measures ANOVA analysis. Results supported the hypotheses that task complexity and goal orientation interact in their effects on performance and satisfaction with performance: the study showed that individuals with a performance goal orientation outperformed those with a learning orientation on a simple task, while no goal orientation effects were observed in the complex task condition; moreover, while individuals with a performance goal orientation were more satisfied with their performance on a simple than on a complex task, those with a learning goal orientation were unaffected by task difficulty. This implies that striving to achieve a tangible/specific goal when facing a complex task results in worse performance and may even turn out to be frustrating, therefore setting more generic targets in line with underlying interests might produce better outcomes. These findings might represent the basis on which to conduct future research on the relation between the optimal degree of goal specificity and task complexity: my assumption is that, being individuals with a performance goal orientation more affected by the complexity of the underlying task, an excessive goal specificity could significantly affect their performance when confronted with difficult and complex tasks. Although this study alone cannot fully support my assertion, it lays the ground for subsequent analysis.
In his 1991 dissertation[iv], Caryn J. Block analyzes the effects of task complexity in goal setting. She starts from the assumption that challenging and specific goals do not lead to performance improvements when tasks are particularly complex - the reason for this being that they inhibit strategy development that is critical to get better results - and she proposes “learning goals” as an alternative type of target to be used in these situations. Her hypothesis is that task complexity moderates the relationship between goals, strategy and performance (challenging and specific goals will have a positive influence on strategy and performance only when task complexity is low) except in the case of learning goals. She conducted her study in a laboratory setting using 134 (57 M, 77 F) undergraduates enrolled in a psychology course and a 2 X 3 experimental design with 2 levels of task complexity (simple/difficult) and 3 levels of goal orientation (performance-challenging/specific; performance-vague; learning-general). Results from her study failed to demonstrate that subjects with learning goals perform better than subjects with either challenging or vague goals when task complexity is high; however, what emerged was that the only the former were able to sustain their level of strategy efficiency when the complexity of the task increases. Although Block’s findings cannot directly support my thesis, nonetheless their implications are relevant: they demonstrated that challenging and specific goals do not lead to improved performance when task complexity is high, and that learning goals are effective in focusing individuals on strategy development. Therefore, the potential exists for an alternative type of goals to be used when faced with complex tasks, and further research should be conducted to define what the optimal structure of these goals is.
De Shon and Alexander (1996) conducted a study to prove that the moderating effect of task complexity on the goal-performance relationship is due to confounding goal difficulty with explicit and implicit learning. Their hypothesis was that setting difficult, specific goals results in improved performance on complex tasks that are processed explicitly[v] while this is suboptimal when implicit learning[vi] is involved. Therefore, to properly study the effects of goal setting on complex task performance, it is necessary to disentangle goal setting and learning mode effects. They considered unsurprising that setting difficult specific goals had generally not improved performance (or even worsen it): complex tasks used in previous research had strong implicit learning components, implying that the effects of learning mode had been so far confounded with the effects of goal setting. To this extent, they conducted two studies to observe the effects of goal setting under the two different learning conditions, on a sample of 124 students enrolled in psychology classes at a large mid-western university. The first study examined complex task performance under conditions which require implicit processing, and results replicated the common finding that goal setting on complex tasks does not result in improved performance relative to “do your best” goals. The second study examined the effects of goal setting when the individual processes the task in an explicit fashion: consistently with the researchers’ hypothesis, results demonstrated that goal setting can have positive effects on performance when an individual process the task explicitly. The results of these two experiments emphasize the need for a careful distinction between implicit and explicit processing when dealing with goal setting on complex task performance, and give important contribution to the hypothesis this paper is addressing: they could be taken, in fact, as evidence supporting my initial prediction, provided that the complex task at stakes involves implicit, rather than explicit, learning processes.
Goals setting research has consistently demonstrated the effectiveness of individual goal setting on task performance, while relatively less is known in relation to group performance. In their study, Kernan et al. (1994) focus on how contextual factors such as task complexity may differentially affect performance in individual and group settings. Recent work[vii] suggested – in line with my initial assumption - that individual goal setting works better on simple rather than on complex tasks, and that assigning specific goals in these cases may result more detrimental then assigning no goals at all. The authors of this study investigated whether the above conclusions are peculiars of individual goals setting, or could be expanded to incorporate group processes as well. Subjects were 180 undergraduate students (83 W, 97 M) enrolled in a eastern college, with mean age 22; for the purpose of the study they were divided in 42 individual and 46 three-person groups. A 2 X 2 X 2 factorial design was used to test the effects of task complexity, information and goal type on performance, with task complexity (high/low), strategy information (high/low) and type of goal setting (individual/group) as independent variables. The results of the study revealed that groups’ performance was largely unaffected by task complexity: teams, in fact, were able to capitalize the advantages of working together, maximizing process gains while at the same time minimizing process losses. Interestingly, these conclusions could imply that, notwithstanding my hypothesis seems to be valid for individuals and applicable in one-man work settings, its legitimacy cannot be expanded to group situations.
Believing that previous research had given little attention to the role of task characteristics as moderators of goal effects, Wood et al. used meta-analysis procedures to assess the effects of task complexity on goal-setting studies conducted from 1966 to 1985: these included 72 studies of goal-difficulty effects (difficult vs. moderate/easy) and 53 of goal specificity-difficulty effects (specific-difficult vs. do-your-best or no goal). The purpose of their study was to examine the moderating effects of task complexity across existing goal-setting research: based on earlier speculations[viii], the authors expected to find larger effects for studies using simple tasks and smaller effects for studies with more complex tasks. Their specific hypothesis was that the positive performance effects of both specific and difficult goals would have been greater on simple than on complex tasks. The results of the analysis provided strong evidence for the hypothesized moderating effects of task complexity on the relationships between the goal attributes and task performance, and task complexity was found to be the only variable to have a significant and robust moderating effect on the performance gains that result from specific, difficult goals. Research should now focus on how goal effects vary as a function of different types of complexity, and on the underlying processes by which goals affect performance on different types of tasks. This study certainly legitimizes my hypothesis that the more complex an activity, the more arbitrary and risky is the setting of very specific goals, and lays the basis to understand how goals should be designed in such circumstances.
An interesting study on the influence of task complexity in goal setting was conducted in 1985 by P. C. Earley. His research focused on the role information play in goal acceptance and performance, and if a relation existed with task complexity. Although previous researchers found that increasing task complexity reduces goal acceptance and performance[ix], it was unclear if providing individuals information could compensate for a complex task, if this information enhanced the individual 's work skills[x]. Earley hypothesized that, although in the case of little task relevant information an individual would outperform and report higher levels of goal acceptance when working on a simple task than a complex one, these differences would not be significant when information about how to perform effectively on the task were provided. His study employed 96 volunteer students (42 male, 54 female) enrolled in social psychology classes, and data were analyzed using a 2 (information provided) X 2 (choice over strategy and work break) X 2 (task complexity) completely crossed, factorial design. Results confirmed previous findings that task complexity was inversely related to goal acceptance and performance, but most importantly demonstrated that giving task-relevant information compensates for an increase in task complexity; the study, however, did not identify what type of task information would be most beneficial. These findings suggest - as an alternative to my proposed solution of assigning more generic goals - to provide more task-relevant information when task complexity is high: this could certainly be a valid possibility, yet it is crucial to specifically understand what kind of information would have the greatest impact.
In conclusion, review of the literature seems to suggest that an unequivocal answer to my question does not exist. Although the mitigating effect of task complexity on performance has been repeatedly proved, it is not clear if this depends on the type of goal orientation, goal’s characteristics, learning mode, group or individual setting or information to which individuals have access. It is manifest that setting difficult and very specific goals cannot work for complex tasks, but it is not clear yet what the best way out to this problem is. Further research needs to be conducted in order to understand how to best set goals when facing task complexity in order to effectively cope with cognitive load, but the common starting point sets an encouraging basis for fruitful future research.

Andrea Balestrino
References & notes • Richard De Shon and Ralph Alexander; “Goal Setting Effects on Implicit and Explicit Learning of Complex Tasks”; Organizational behavior and Human decision process; Vol. 65(1); 1996. • Debra Steele-Johnson et al.; “Goal Orientation and Task Demand Effects on Motivation, Affect, and Performance”; Journal of Applied Psychology; Vol. 85, No. 5; 2000. • Caryn J. Block; “A contingency approach to goal setting: taking task complexity into account”; Dissertation (NYU); 1991 • Marie C. Kernan et al.; “Individual and group performance: the effects of task complexity and information”; Human performance; Vol. 7(4), 1994. • Robert E. Wood et al.; “Task Complexity as a Moderator of Goal Effects: A Meta-Analysis”; Journal of Applied Psychology; Vol. 72(3); 1987. • P. Christopher Earley; “Influence of Information, Choice and Task Complexity upon Goal Acceptance, Performance, and Personal Goals”; Journal of Applied Psychology; Vol. 70(3); 1985.

-----------------------
[i] A learning goal orientation cues an individual to believe competence can be improved, to evaluate competence in relation to previous competence, and to choose and persist on a challenging task

[ii] A performance goal orientation cues an individual to believe competence is not likely to change, to evaluate his competence in relation to others, and to choose a task in which he can prove his competence and avoid failure

[iii] Dweck, 1986; Farr, Hofmann, & Ringenbach, 199; Button et al., 1996; Vlachopoulos & Biddle, 1997

[iv] Department of Psychology, Graduate School of Arts and Science - NYU

[v] Explicit learning: the individual consciously develops a mental representation of the problem, formulates strategies, and test alternative hypotheses for task performance

[vi] Implicit learning: the individual learns complex rules of system behavior without conscious attempt to do so

[vii] Earley 1985; Huber, 1985; Jackson and Zedek, 1982; Wood, Mento et Locke, 1987

[viii] Locke et al., 1981; Wood, 1985

[ix] Steers & Porter,1974

[x] Beehr & Love, 1983

References: [vi] Implicit learning: the individual learns complex rules of system behavior without conscious attempt to do so [vii] Earley 1985; Huber, 1985; Jackson and Zedek, 1982; Wood, Mento et Locke, 1987 [viii] Locke et al., 1981; Wood, 1985 [ix] Steers & Porter,1974 [x] Beehr & Love, 1983