Strategic adaptability in negotiation: a framework to distinguish strategic adaptable behaviors

Purpose – This study aims to propose and evaluate a novel framework of strategic adaptability in dyadic negotiations. The authors de ﬁ ne strategic adaptability as a reaction to a cue that leads to shifts between integrative and distributive strategies. Based on the literature on turning points, phase models and strategic negotiations, theauthors developed aninitialframework identifying ﬁ ve distinctstrategic adaptations. Design/methodology/approach – To verify the framework, the authors analyzed two negotiation simulationswithadiversesetofnegotiationstudents.Negotiationswerecontent-coded,andadaptationswerelabeled. Findings – The authors found a consistent pattern across two studies. Overall, 12% (study 1) and 18% (study 2) of all speaking turns were identi ﬁ ed as strategic adaptations. The ﬁ ndings empirically con ﬁ rmed four of their strategic adaptation types: adapt to deadlock, follow adaptation by opponent, adapt to priority of issue under discussion and adapt to new information on issue. Moreover, ﬁ ndings of this study revealed two new types of strategic adaptability: delayed adaptation to opponent and adapt to understand opponent. Study 2 additionally revealed that strategies vary with the negotiation phase

Negotiation is an adaptive process influenced not only by its context but also by the strategic choices of the opponent (Olekalns and Weingart, 2003).Because of its inherent uncertain and dynamic nature, the ability to react and adjust strategies to the unfolding negotiation process effectively is key to negotiation success (Hawes and Fleming, 2014).
In negotiation, strategic adaptability has been coined as the ability to purposefully adjust to evolving processes by changing own actions or approaches (Martin et al., 2012;Smolinski and Kesting, 2013).Specifically, adaptable negotiators swift smoothly between more cooperative, integrative behaviors and more competitive, distributive behaviors when needed (Smolinski and Xiong, 2020).However, studies indicate that such adaptations do not come naturally because it is challenging to apply both behaviors throughout the negotiation process and switch between them at the right moment (Thompson, 2009;Chapman et al., 2017).
Considering that strategic adaptability is an essential skill for negotiators, surprisingly little research has been done to conceptualize and empirically investigate strategic adaptability.This is remarkable since the literature and textbooks stipulated the importance of combining integrative and distributive strategies decades ago (Walton and McKersie, 1965;Lax and Sebenius, 1986;Pruitt and Rubin, 1986).This study is, to our knowledge, the first attempt to empirically investigate the concept of strategic adaptability in two ways.First, we aim to develop a framework including categories and cues of strategic adaptability in the negotiation process by integrating insights from research studying significant shifts in negotiation dynamics, such as work on turning points (Druckman, 2001) and phase models (Weingart and Olekalns, 2004).We also draw upon work departing from a strategic contingency perspective to negotiate, highlighting the specific issue under consideration and how adaptation may occur because of the interpretive frame and motivations behind a topic (Walton et al., 1994;Pruitt et al., 2004).
Second, we put this framework to a first test by analyzing the behavioral dynamics in two samples of negotiation simulations.In study 1, we introduce a stepwise method to capture those moments of strategic adaptation and investigate their specific features.This results in the first framework of categories of strategic adaptability, which we further validated in a follow-up study (study 2) in a more complex negotiation setting with trained participants.The results of our analysis set a research agenda for future scholarly work on strategic adaptability and explore how negotiators can adapt their negotiation strategies in response to each other.

Integrative and distributive negotiation
Negotiation is a process that occurs because the involved parties believe that an agreement on conflicting objectives is a necessary means to attain their goals (Lewicki et al., 1999).Because of its inherent mixed-motive character, negotiators need to balance between cooperative, integrative strategies and competitive, distributive strategies (Walton and McKersie, 1965;Brett, 2000;Beersma and De Dreu, 2002;Fulmer and Barry, 2004;Saorín-Iborra and Cubillo, 2019;Steinel and Harinck, 2020).Although research shows that a IJCMA mixed-strategy approach leads to better outcomes than either one alone (Van de Vliert et al., 1999), doing so does not come easy.An important reason for this is that the behaviors encompassing each strategy tend to differ considerably and psychologically represent opposite directions.Integrative strategies focus on finding a mutually beneficial agreement that satisfies the needs of both parties.The parties try to find common ground and explore multiple options to reach a win-win solution.Integrative bargaining tactics seek to create joint value by exploring mutual interests, sharing information openly, joint problem-solving, incorporating multiple issues simultaneously and implementing creative approaches.This often requires open communication, willingness to compromise and a focus on long-term relationships (Weingart et al., 1990;Fulmer and Barry, 2004).
In contrast, distributive strategies focus on winning and maximizing one's own gains, often at the expense of the other party (Fulmer and Barry, 2004).As a result, distributive bargaining tactics are often adversarial, with each party trying to gain an advantage over the other.Tactics such as making extreme demands, using threats and persuasion are common in distributive bargaining (Zartman, 1977;Barry and Friedman, 1998).
A critical difference between the two bargaining strategies is the use of threats (De Dreu et al., 2000).In distributive bargaining, threats and ultimatums are commonly used.These threats can be real, such as the threat of walking away from the negotiation, or they can be bluffing tactics ( Van de Vliert and De Dreu, 1994).In comparison, integrative bargaining strategies use a more collaborative approach to find mutually beneficial solutions [Table 1 presents an overview of integrative and distributive behaviors based on e.g., Giebels et al. (1998) and Weingart et al. (2004)].A detailed overview of the behaviors, including definitions and examples, can be found in Appendix 1.
To use an effective mixed-strategy approach, negotiators not only have to be able to apply both strategies but also be strategically adaptable to switch between them and recognize at what moment they should do so.

Strategic adaptability in negotiation
Adaptability can be defined as the ability to purposefully adjust to changing circumstances by changing own actions or approaches to fit different environments and conditions (Martin et al., 2012).Applying this definition in the negotiation context, we define strategic adaptability as 1) a reaction to an informational cue, 2) requiring change from a more distributive strategy to a more integrative strategy, or vice versa.
In what follows, we elaborate on the different informational cues encompassing strategic adaptability.

Strategic adaptability in negotiation
An appropriate point of departure for understanding adaptation processes in negotiation is to examine the literature focusing on substantial shifts in negotiation dynamics.Such changes likely contain clues to strategic adaptability.

Turning points
An important line of research in this respect focuses on turning points, referring to transitions from one consistent set of movements or strategic orientation to another (Druckman and Olekalns, 2011).For example, the change consists of moving away from a give-and-take pattern to offering a mutually beneficial solution instead (Druckman, 2001;Druckman and Olekalns, 2011) or sharing new information that changes the perspective on issues in the negotiation (Druckman, 2001).
The literature on turning points identifies three categories of events leading to a turning point in negotiations.One category refers to external events, and two others refer to internal events that occur inside the negotiation.External turning points occur outside the negotiation, such as policy or leadership changes or the inclusion of third parties (Crump and Druckman, 2012).They influence the negotiation context and, therefore, arguably the perceptions or priorities of the involved parties.Such shifting perceptions or priorities may serve as a cue to strategically adapt (context adaptability).
Internal turning points that directly relate to the negotiation process can be caused by process cues.Those process cues may trigger decisions to change the structure or format of the negotiation.A typical cue would involve a crisis that threatens the continuation of the negotiation, such as a deadlock (Druckman, 2001).Particularly, adapting to a deadlock can reflect a moment of strategic adaptability because they refer to still points in negotiation when no progress seems possible.An indication of such a strategic adaptation in an ongoing negotiation might be that one of the negotiators would highlight the deadlock in a metaconversation and suggest a way out.As a breakthrough moment is required to lift the blockade and continue the negotiation (Druckman, 2001), it may serve as a cue to strategically adapt (process adaptability).
Another internal turning point can be caused by substantive cues.Substantive cues refer to both new ideas (e.g., progress seeking) or information that delinks or reframes issues.Such cues lead to new framework agreements or ways of packaging proposals (e.g., logrolling) (Druckman, 2001).Therefore, substantive cues can lead to a change in the negotiation process (Druckman and Olekalns, 2013), which requires strategic adaptation to new insights or ways of thinking about issues (content adaptability).

Strategic negotiation
Similarly to the turning point literature, research on strategic negotiation is associated with content adaptability and adds to our understanding of strategic adaptability.In line with previous work on mixed-motives (Deutsch, 1958(Deutsch, , 1973;;Pruitt and Carnevale, 1993), Walton et al. (1994) influential work on strategic negotiations explained conditions that promote distributive and integrative strategies.Given that most negotiations involve multiple issues and negotiators usually do not necessarily attain the same weight to each issue, negotiators are likely to adapt their strategy to the specific issue under discussion.For instance, objectives such as substantive goals with high priority are more likely to promote a distributive strategy.In contrast, social goals with high priority promote integrative behavior (Walton et al., 1994).Therefore, negotiators need to be both firm and flexible (Pruitt, 1981;Pruitt and Carnevale, 1993), depending on the circumstances that are likely to change during negotiation.The effectiveness of different negotiation behaviors depends on the contingencies and IJCMA circumstances (Axelrod, 1984;Van de Vliert et al., 1999).Specifically, the applied strategy depends on the perceived priority of the issues under discussion and the feasibility of attaining them.As such, a specific category of content adaptation emerges tailored to the issue under discussion and based on cues such as a change of topic or receiving a concrete offer (e.g., logrolling).

Phase models
Lastly, the research on phase models is relevant to our inquiry and distinguishes between stage and episodic models.Stage models divide negotiations into fixed intervals and investigate strategies across each phase (Olekalns et al., 2003).According to this model (Holmes, 1992), negotiators use different strategies at the beginning, middle and end of a negotiation (Olekalns et al., 1996;Preuss and van der Wijst, 2017).
In contrast, episodic models focus on consistent periods of coherent patterns of behaviors (i.e., strategy sequences) that define a phase (Baxter, 1982;Olekalns et al., 2003;Weingart and Olekalns, 2004).Throughout the negotiation process, negotiators respond to one another, and the frequency of integrative or distributive strategies aggregates over time.This aggregation of strategies is named strategic sequences (Olekalns and Weingart, 2008).An initiated shift in such a pattern, caused by a change in the opponent's behavior (e.g., distributive behavior after a sequence of joint integrative ones), may serve as a cue to adapt negotiation strategies.At that moment, negotiators need to evaluate if it is in their interest to maintain or redirect their behaviors (Olekalns and Weingart, 2003).
Episodic model research shows three types of sequences distinguishing between maintaining and redirecting a strategy (Putnam and Jones, 1982;Donohue et al., 1984;Olekalns and Smith, 2000).The first two involve maintaining the dominant strategy by either reciprocation of the same behavior (reciprocal sequences) or with other behavior falling within the dominant (distributive or integrative) strategy (complementary sequences).The third category, transformational sequences, involves moving from integrative to distributive patterns or vice versa.Such mismatching of strategies provides a clue that a strategic adaptation is taking place.Focusing on the action-reaction nature of sequences (i.e., negotiators' responses) may thus provide insights into how the shifting behavior of party A (i.e., the initiator) is the cue for the strategic adaptation of party B (i.e., the follower).Based on the change in strategy from the initiator (which could be an act of strategic adaptability as well), the opponent has to decide whether or not to follow the change in behavior of the opponent (opponent adaptability).An example would be a tit-fortat strategy where negotiator B reciprocates the change to a distributive strategy of negotiator A on the previous move (Axelrod, 1984).
Taken together, the literature shows that there are several reasons to strategically adapt one's behavior.When examining these different reasons, we observe that they occur on different levels.Table 2 presents the research lines and connects them to four overarching categories that involve changes on the context, process, content and opponent levels.Furthermore, the categories encompass five types of strategic adaptations in negotiations and build the basis for the framework we will test in this paper.

Study 1
In study 1, we present the results of a negotiation simulation study and aim to answer: RQ1. Can we capture the moments of strategic adaptability as depicted in Table 2 and meaningfully distinguish between those categories?

Strategic adaptability in negotiation
Categories  During the training, the participants received an introductory lecture on negotiation, practiced negotiations based on a multi-party negotiation simulation, and partook in a twoparty simulation, which we used for this study.We obtained consent from all participants except for one whose negotiation dyad was excluded from further analyses.Both studies were jointly approved by the ethical committee of the authors' faculty.The remaining sample of 66 participants had an average age of 20.6 years (SD ¼ 1.61), and 68% were male.Most participants were Dutch (76%) and German (13%).The remaining participants came from other countries in Europe (6%), Asia (3%) and Africa (2%).
Procedure.Due to COVID-19 regulations, the negotiation was performed online using Microsoft Teams' videoconferencing platform.The participants received descriptions of the negotiation simulation 30 min before the negotiation started and were given instructions to use the time to prepare for the upcoming negotiation.In all negotiations, both participants switched on their cameras for face-to-face negotiation.The average negotiation time was 16.4 min (SD ¼ 11).After the simulation, the participants reported their agreed outcomes.
The participants undertook a prototypical role-play exercise often used during negotiation training (Lewicki, 1997), using a scenario suggested by Thompson (2009), which we adapted to fit this research.The scenario (see Appendix 2 for the complete summary) was changed by adjusting the setting to a familiar environment to increase realism.All participants received a description of the conflict from their perspective, which was not allowed to be shared.The simulation was designed so that it allowed for both integrative and distributive negotiation.The administrator randomly paired up the participants.The negotiation task can be summarized as follows: Two participants that are co-tenants of an apartment have to resolve a conflict on a late charge fee and duties in the apartment while maintaining the relationship.Both sides believe they should not pay the late charge fee.Both sides are busy with their studies and personal lives and dissatisfied with the relationship as co-tenants.To improve the relationship, they could propose ideas on how to divide the task in the house and activities they can do together.Therefore, the instructions on both sides included an activity that could be suggested.
Response coding.In total, 540 min of video recordings were transcribed, resulting in 1,625 speaking turns.The content of the transcribed negotiations was analyzed following previously documented process analyses (Donohue et al., 1984;Weingart et al., 1996;Brett et al., 1998).Each speaking turn was coded in the first phase on different types of integrative and distributive strategies, followed by the second phase, where the adaptation in strategy was coded based on the categories in Table 1.
First phase.A coding scheme was developed using a deductive approach, distinguishing between integrative and distributive strategy codes (see Appendix 1 for more details).Similar to Olekalns et al. (2003) and Brett et al. (1998), we used two coders: an independent coder, blind to the negotiated agreements and the research questions, and the first author, coded the data to determine the inter-rater reliability (McHugh, 2012).Each speaking turn received a code using the qualitative data analysis software NVivo (QIP Ltd, 2012).The integrative and distributive strategy codes attained a strong inter-rater agreement (Cohen's k ¼ 0.90).The remaining disagreements were resolved through discussion and consensus, which was necessary to continue the coding of strategic adaptations (second phase).Of the participants, 7% used only distributive, 29% only integrative and 64% used a mix of both strategies (i.e., adapting strategy at least once).

Strategic adaptability in negotiation
Second phase.A second coding scheme was developed, in which we used the adaptation categories found in Table 2 to identify the types of adaptations that were observed in the negotiation process.Hence, we used a deductive approach in the second phase of coding (Hyde, 2000).The same coders analyzed 188 of the strategic adaptations (12% of all speaking turns) with this coding scheme.A sufficient inter-rater agreement (Cohen's k ¼ 0.72) was attained.The coders resolved the disagreements through discussion.During the coding, not all adaptations could be attributed to the five predetermined categories, indicating that the categories are incomplete.We, therefore, further analyzed the unattributed texts using inductive reasoning, starting with observations to establish generalizations (Hyde, 2000) [1].

Results
Table 3 demonstrates the strategic adaptability categories investigated in a negotiation, where five categories stem from the literature and two emerged from the data.The categories are grouped into context-, process-, opponent-and content-based adaptations.The context-based adaptation, adapt to external factors, was not observed in this research.However, two new categories of adaptations emerged in the overarching category of Opponent-based adaptations.We termed the first new category delayed adaptation to opponent.This code occurs when a strategic adaptation from party A (shift from distributive to integrative, or vice versa) is not directly followed by a shift in strategy from party B but in a delayed fashion after two or more speaking turns.Alternatively, this code also occurs when party A adapts the strategy and party B does not follow the change.As a response, party A changes back to the initial strategic orientation to synchronize the strategy with party B after one or more speaking turn (i.e., a delayed adaptation from party A).In addition, we observed a second new adaptation category termed adapt to understand opponent, which refers to a negotiator adapting to understand the underlying concerns of the opponent.Specifically, this code refers to situations where the negotiator adapts their strategy to understand the opponent or seeks to clarify their opponent's interests, concerns, feelings, motivations-or thoughts.This would be the case, for example, when party A shares feelings and concerns, and party B would consider this a cue to switch from a distributive strategy to an integrative one by asking open-ended questions, acknowledging the opponent's concerns or summarizing information.Table 3 summarizes the results of study 1 and describes the observations for each adaptation individually.For a comprehensive overview of all strategic adaptations, we included the context category (adapt to external factors).However, this category was excluded from further analysis because it was outside the scope of study 1.
Table 4 shows the total percentages of all observed strategic adaptations and the percentage of integrative and distributive strategies.Most strategic adaptations were observed for the codes adapt to priority issue under discussion, follow adaptation by opponent and delayed adaptation to opponent.Those codes have almost an equal distribution between integrative and distributive adaptations.Interestingly, the other three codes of adapt to understand opponent, adapt to deadlock and adapt to new information on issue were less frequently observed, and adaptations only occurred from distributive to integrative strategies.

Discussion study 1
We established seven categories of strategic adaptations (12% of all speaking turns).Four of them, adapt to priority issue under discussion, follow adaptation by opponent, adapt to new information on issue and adapt to deadlock, were derived from the literature.One categoryadapt to external factorswas not observed, presumably due to the nature of the task (a IJCMA short, straightforward negotiation roleplay without external influences).Interestingly, we also observed two additional categories of strategic adaptability: delayed adaption to opponent and adapt to understand opponent.Both can be considered Opponent-based adaptations because the opponent's behavior likely triggers the adaptation.The delayed adaptation to opponent category refers to a short period of asynchronized strategies before following the strategy of the opponent (e.g., a delayed tit-for-tat strategy or due to the opponent's Adapting to an opponent's change in strategy with a delay or adapting to the opponent's resistance to follow a change in strategy Negotiators A and B are using similar strategies.When negotiator A changes to an integrative or distributive strategy, negotiator B does not directly synchronize the chosen strategy but instead delays its adaptation.Alternatively, Negotiator B does not follow, and A changes back to the initial strategic orientation to synchronize the strategy Adapt to understand opponent c Adapting the strategy to understand or clarify the opponent's interests, concerns, feelings, motivation or thoughts Negotiators A and B can have (a) synchronized strategies.Negotiator A shares feelings, concerns or makes an offer, and negotiator B responds by adapting its strategy to understand A's concerns (e.g., active listening, acknowledging feelings or summarizing information) Notes: a The Context category was not observed in study 1 or study 2; b Even though we represent our results chronologically, this code was added after the analysis of study 2 because of the infrequent occurrence in study 1; c Newly added categories introduced following the analysis of studies 1 and 2 Source: Authors' own work

Strategic adaptability in negotiation
resistance to follow the change in strategy).The adapt to understand opponent refers to adapting the strategy to understand the underlying concerns of the opponent.This category parallels the concept of active listening, referring to constructive listening while focusing on understanding the opponents' feelings or empathizing with them (Cambria et al., 2002).Furthermore, we were able to give some first insights into the relative occurrence of each category of strategic adaptations, with the content-based category adapt to priority issue under discussion being most frequently used.Our analyses also provided fine-grained examples of each strategic adaptation category and the cues that might have led to them.These findings are promising, as they clearly indicate that moments of strategic adaptability can be captured and meaningfully distinguished between.To increase the robustness of these findings, we conducted a follow-up study with trained participants.

Study 2
Considering the strategic adaptation categories identified in study 1, the goal of study 2 was to replicate the findings of study 1 and build upon it in three ways.First, we selected a group of participants that have been trained in negotiation before to explore if the same strategic adaptations occur with trained negotiators in another simulation.We aimed to answer: RQ2.Are the same strategic adaptations observed in another simulation with trained negotiators?
Second, we examined whether adaptations vary in different phases of the negotiation process.According to the stage model (Holmes, 1992), negotiators use different strategies at the beginning, middle and end of a negotiation (Olekalns et al., 1996; Preuss and van der Wijst, 2017).We followed Holmes (1992), who divided negotiations into fixed time intervals and explored how adaptations differ across three phases.We specifically aimed to answer: RQ3.What commonalities or differences can be observed regarding the occurrence of strategic adaptations across the three negotiation phases?
Third, we examined whether strategic adaptability relates to improved individual outcomes or improved quality of joint outcomes.The literature indicates that combining integrative

Participants
The participants were 64 students following extensive negotiation courses at a university in The Netherlands, Japan or Iceland.We obtained consent from all participants except for two whose negotiation dyads were excluded from further analyses.These 60 participants had an average age of 29.4 years (SD ¼ 8.27), and 55% were male.The majority of the participants came from Iceland (45%) and countries from Asia (32%), Europe (15%), Africa (3%), South America (2%) and North America (3%).

Procedure
The participants received a negotiation simulation via e-mail to prepare and perform.The students were randomly paired with a participant from another university to ensure that negotiation dyads were unacquainted.Zoom software was used as the negotiation platform and the negotiation was completed in one session.The average negotiation time was 43 min (SD ¼ 24.8 min).In addition, the negotiations were recorded, transcribed and coded based on the categories of strategic adaptability from study 1.After the simulation, the participants received a post-questionnaire, including a five-item questionnaire on the final agreement to calculate the individual outcome and quality of their joint outcome.

Negotiation role-play
A two-party, multi-issue negotiation simulation was adapted from Giebels et al. (1998) and Thompson (1990a).Each participant received a description of the negotiation scenario from their role's perspective, which was not allowed to be shared.The negotiation was between a purchasing and financial manager from the same organization regarding the purchase of new software.Five issues (purchasing software, time to launch, duration of workshops, maintenance and service fee and project budget) needed to be agreed on while maintaining the relationship.Each issue had a different profit score (see Appendix 3), allowing for trading priority issues to generate value for both sides.Instructions were added that offered each side the opportunity of a bonus and consequences if no deal was reached.The two parties were working for the same organization, which allowed them to focus both on the organization's goal and on individual gains, which allowed them to use both integrative and distributive behaviors.

Response coding
In total, 1,285 minutes of video recordings were transcribed, with 2,417 speaking turns.As in study 1, each speaking turn was coded in the first phase on integrative and distributive strategies (see Appendix 1).An independent coder, blind to the negotiated agreements and the research questions, and the first author coded the data to determine the inter-rater reliability (McHugh, 2012).The integrative and distributive strategy codes attained a good inter-rater reliability (Cohen's k ¼ 0.82).The remaining disagreements were resolved through discussion and consensus in a similar manner as in study 1.The same coders continued in a second phase with the coding of a total of 447 strategic adaptations (18% of all speaking turns).The coding categories were defined in study 1 and are based on Table 3.

Strategic adaptability in negotiation
A strong inter-rater agreement (Cohen's k ¼ 0.91) was attained.The remaining disagreements were resolved through discussion.Of the participants, none used only distributive, 8% only integrative and 92% used a mix of both strategies (e.g., adapting strategy at least once).Hence, compared to study 1, study 2 showed a higher percentage of using a mixed strategy approach.

Strategic adaptation categories
Table 4 shows the total percentages of all observed strategic adaptations and percentages of integrative and distributive strategies.The results of study 2 show a highly similar pattern to the results of study 1.Equivalent to study 1, most strategic adaptations were observed in the overarching categories of Content-and Opponent-based adaptations.The two codes of follow adaptation by opponent, and delayed adaptation to opponent, have almost an equal distribution between adaptations from integrative and distributive strategies.Furthermore, the two codes adapt to understand opponent and adapt to deadlock were observed more frequently compared to study 1.Both codes only occurred from distributive to integrative strategies, as in study 1.Interestingly, the code adapt to priority issue under discussion occurred more frequently from integrative to distributive adaptations than in study 1.The overall use of strategic adaptations seems somewhat higher in study 2 (18% of all speaking turns), which is likely due to the participants having more negotiation experience and the role-play including a greater number of issues to be agreed on.

Distribution of strategic adaptations in time
To examine whether adaptations differ across phases, we divided each negotiation into three equal phases of length (i.e., three times 1/3 of the negotiation time) (Holmes, 1992).Table 5 illustrates the observed strategic adaptation categories per phase.Considering the total frequency of each category, there appears to be a clear pattern with two dominant categories, adapt to priority issue under discussion and follow adaptation of opponent, being most prominent in all phases.Only adapt to new information on issue is almost not observed in the third phase, which seems to be used more frequently at the beginning.While we observe a steady increase over time in percentages of strategic adaptations for the categories follow adaptation by opponent, delayed adaptation to opponent and adapt to priority issues Notes: a The focus lies on the columns.The sample size of negotiators using both integrative and distributive strategies is N ¼ 55.The relative frequency distribution of strategic adaptations over three phases (T) in the negotiation is used to measure the distribution of adaptations.Therefore, the negotiations were divided into three equal parts.The relative frequencies of total adaptations observed in study 2 per phase are T1 ¼ 111; T2 ¼ 145; T3 ¼ 191 Source: Authors' own work IJCMA under discussion.Therefore, the findings indicate that, on average most of these adaptations occur toward the end of the negotiation [2].Adapt to understand opponent shows a different pattern where most adaptations occurred in phase 1 and phase 3, which suggests that negotiators adapt their strategy to resolve unclarities mainly at the beginning and end of the negotiation.Not surprisingly, adapt to deadlock occurred most frequently in the last phase.

Adaptations and outcomes
In Table 6, we present the outcome of each individual participant [3].Each participant could attain a maximum individual outcome of 1,120 points and a maximum joint outcome of 1,560 points if they were able to resolve all issues.We examined whether strategic adaptations influence individual and joint outcomes.The joint outcomes were determined by whether the participant negotiated a pareto optimal agreement [e.g., distributes resources without decreasing the agreement's value for the opponent (Baumol, 1977, p. 527;Coleman, 1979;Barr, 2012)].We divided participants into four categories according to the outcome they attained within the dyads.These categories group dyads into: Although we realize individual outcomes are nested in dyads and findings should therefore be interpreted with caution, Table 6 provides some interesting insights.Table 6 illustrates the findings of differences in strategic adaptability between the four outcome groups, including some demographic variables.Interestingly, negotiators able to attain a Pareto-optimum outcome used strategic adaptations the least, while negotiators who scored worse relative to the other side (the low-high group) used them the most.These lower-scoring negotiators seem to adapt to the opponent's strategy more frequently instead of initiating a change in strategy.Furthermore, lower-scoring dyads (the low-low group) used more adaptations due to a deadlock, implying that this group experienced more still points where no progress seemed possible.Instead, higher-scoring negotiators (the high-low group) used more adaptations to priority issues under discussion, which indicates that those negotiators led the discussion by changing the strategy based on the topic.Furthermore, the findings show that all four outcome groups used the most adaptations in the last phase.However, higher-scoring negotiators used only 43% of adaptations compared to the lowerscoring dyads, who used the most adaptations (80%).

General discussion
Many scholars refer to strategic adaptability as an essential skill for negotiators and emphasize the cumulative use of integrative and distributive strategies (Walton and McKersie, 1965;Lax and Sebenius, 1986;Pruitt and Rubin, 1986;Martin et al., 2012).However, the literature provides little insight into how exactly negotiators adapt strategies in the negotiation process and provides some anecdotal but not comprehensive empirical evidence.As a consequence, we know little about the cues signifying change, dominant behaviors for each adaptation type, and how strategic adaptations unfold during the negotiation process.Our research seeks to conceptualize strategic adaptability, understand how negotiators adapt in the negotiation process and identify cues leading to change.

Theoretical contributions and implications
Point of departure for our inquiry was previous work on turning points, phase models and strategic negotiations which we integrated in an initial framework.In two studies, we put this framework to the test, expanded it and found some consistent patterns.The final framework includes a comprehensive set of seven different types of strategic adaptations.Five categories and cues of strategic adaptability are based on the literature, with two additional ones emerging from our data (delayed adaption to opponent and adapt to understand opponent).
The framework suggests concrete cues and examples for each category.Generally, our adaptation categories show how negotiators can react and change to the context, process, content and opponent.This information can be of strategic value and adds to the literature (Weingart et al., 1990;Olekalns and Smith, 2000;Olekalns et al., 2003;Olekalns and Smith, 2003) because it informs us about how negotiators initiate change and the underlying intentions when strategically adapting.Both studies show a consistent pattern where most IJCMA strategic adaptations occurred in the overarching categories of Content-and Opponent-based adaptations.Furthermore, the adaptations occurred in a similar pattern to integrative and distributive strategies.Overall, we can establish that it is important to investigate strategic adaptations further, as they were used between 12% and 18% in the negotiation of all speaking turns.
Second, we found two new types of adaptations previously not described in the negotiation literature (Walton et al., 1994;Druckman, 2001;Pruitt et al., 2004;Weingart and Olekalns, 2004;Olekalns and Weingart, 2008): delayed adaptation to opponent and adapt to understand opponent.Delayed adaptation to the opponent indicates that negotiators might not directly identify the shift in the opponent's strategy and want to keep the strategy the same or do not want to change at all.Adapting to understand the opponent indicates that if a negotiator shares concerns and feelings, the distributive-oriented opponent might change the strategy in response.Although more in-depth research is required to understand the underlying mechanisms behind these new adaptation categories, they propose additional cues and allow for a more precise examination of reasons that lead to a change in strategy.
Third, our findings suggest that different types of adaptations might be more suitable at different negotiation phases and change over time.These insights are relevant because they provide more detailed information on how negotiation dynamics change during the negotiation.Depending on the type of adaptation, specific cues can be used to influence the negotiation process and change the strategic orientation.Our findings suggest that certain types of adaptations only or mainly encompass changes from distributive to integrative strategies (e.g., adaptations to a deadlock, to understand the opponent and to new information on the issue).In contrast, others seem to occur more frequently towards the end of the negotiation (e.g., adaptations to follow the opponent, delayed adaptations and adaptations to priority issues under discussion).These findings indicate why negotiators adapt strategies towards the beginning, middle or end of the negotiation, adding to previous research on negotiation behaviors and strategies (e.g., Brett, 2007, Brett et al., 2004, Donohue, 2004, Butt et al., 2005, Olekalns and Weingart, 2008;Druckman and Olekalns, 2011).
Fourth, although our data only provides preliminary insights that should be interpreted with care, interesting patterns emerged that could explain how strategic adaptability relates to the negotiated outcome.According to these data, we can infer that negotiators adapting their strategy moderately attain high joint outcomes.Whereas more frequent adaptations can lead to negotiation instability, resulting in lower outcomes for at least one party.Our results relate to previous research (Olekalns and Smith, 2000;Hawes and Fleming, 2014;Brett et al., 2021), showing that combining integrative and distributive behaviors can benefit joint outcomes.
Implications for practice.Our research sheds light on the dynamic nature of negotiation and provides a framework for negotiators to consciously adapt their strategy based on specific cues.The framework identifies seven categories of adaptation with specific cues that negotiators can use to change their strategic orientation according to the context, process, opponent and content of the negotiation, highlighting the importance of flexibility and situational awareness.By creating awareness of strategic adaptations, this framework can be used to train negotiators on how to make use of strategic directions and to be aware of the actions of the opponent.The results also show that different types of adaptations might be more suitable at different negotiation phases.Our findings have practical implications for negotiators and educators involved in negotiation and conflict management, providing guidance on how to approach the negotiation process and achieve better joint outcomes.

Limitations
Our study has limitations that should be taken into consideration when interpreting the results.First, the sample size in both studies was relatively small.Therefore, caution should be taken when generalizing our findings.In future research, a more diverse sample in terms of cultural backgrounds and age range should be recruited to enhance the external validity and generalizability of the findings.Second, our coding analysis was conducted by two coders per study.While we ensured inter-rater reliability, having more coders could have strengthened the analysis and increased the validity of the results.Third, we report on integrative and distributive negotiation strategies in general and derived the negotiator's strategies from behavioral observations but did not actually capture the intentions and underlying cognitive processes of the negotiators.Whereas strategy typically refers to intentional behavior (Pruitt and Carnevale, 1993), future research should include cognitive aspects to explore intentions and awareness of changing strategic orientations.Specifically, an information processing approach could be considered, which establishes theories of the mind of the individual negotiator and assesses the cognitive process of judgment (Bazerman and Carroll, 1987;Thompson, 1990b).While negotiators interact with each other, they are faced with complex decision-making tasks where they have to gather and share information, evaluate it with alternative information and make behavioral choices (Bazerman and Carroll, 1987).Given that there is a cue to adapt the strategy, negotiators have to be able to perceive and process the information before they can actually do so.
Fourth, we developed a strategic adaptability framework based on the analysis of two quite similar negotiation simulations.Although the identified adaptation categories align with the more general negotiation literature, it could be that other categories exist that were not captured in our research.For instance, external cues were not observed in our study but are mentioned in the literature (Druckman and Olekalns, 2011) or when both negotiating parties consistently use integrative strategies, a negotiator might decide to adapt to a more distributive style in an attempt to explore the deal's full potential.Future studies could analyze different negotiation settings and apply our framework with negotiation professionals in organizations to examine if our framework could be expanded with additional adaptation categories.

Suggestions for future research
Our findings open up many opportunities for future research.We establish four themes to enhance: the strategic adaptability framework, our knowledge of individual factors, the outcome and education with relevant questions.They are pertinent to address because they help strengthen our understanding of strategic adaptability, its effects on the negotiation process and outcome, and how to educate negotiation students and practitioners on becoming more adaptable.
First, we think it is important to further test the robustness and underpinnings of our framework.This inquiry could concentrate on the following questions: Did we capture all strategic adaptability types in our framework, or can other negotiation contexts add to the framework?For instance, we did not observe external cues in our current research setting.Therefore, can we observe the same strategic adaptations in real-world negotiations?Are strategic adaptations occurring intuitive or conscious when negotiators combine integrative and distributive strategies to achieve their goals and reach an agreement?We also encourage further examination of the cognitive aspects of strategic adaptations.By exploring the intentions and awareness of changing strategic orientations, we could increase our understanding of why negotiators sometimes delay adaptations to the opponent (e.g., do IJCMA they not recognize the shift, or do they not want to shift immediately).Furthermore, it would be insightful for other studies to report on the frequency of using and adapting between integrative and distributive strategies in various negotiation settings to grasp what an average and optimum number of adaptations entails.
Second, further work is required to investigate potential individual, and cultural conditions influencing adaptability (Giacomantonio et al., 2020;Brett et al., 2021).For example, a negotiator's social value orientation (Van Lange, 1992) might impact a preference for certain types of strategic adaptations or how they are initiated (e.g., are pro-self-oriented negotiators inclined to initiate adaptations from integrative to distributive strategies).Similarly, individual differences in personality, such as levels of openness to experience, emotional intelligence, cognitive flexibility and power dynamics, potentially impact strategic adaptability.Specifically, cultural differences have been shown to affect intercultural negotiations (Brett, 2000;Adair, 2003).Therefore, some cultural clusters (Ronen and Shenkar, 2013) might be more open to being strategically adaptable during negotiations.Our sample sizes do not allow for precise statistical analysis of differences within the dominant cultural clusters.Future research could investigate this in more depth and explicitly examine the impact of culture on strategic adaptability.
Third, our data provide preliminary insights into the influence of strategic adaptability on the outcome.Further studies that strengthen our understanding are suggested, for instance: how do the strategic adaptability categories contribute to more favorable negotiation outcomes (e.g., how often should a negotiator adapt and in what order)?What influence have the strategic adaptations, such as adapt to understand opponent or adapt to issue under discussion on building trust, relationships or improving communication?These topics also link to issues of trust in negotiation (Caspi et al., 2017;Druckman and Harinck, 2022).In general, it would be good to include different types of substantive and relational outcomes.
Fourth, to train students and practitioners to become more adaptable to the unfolding negotiation process, educational interventions could be developed.Inspired by recent work on mindset and negotiation training (Ade et al., 2018;Schuster et al., 2022), future work might explore how negotiators can become mindful of strategic adaptations and what tools or techniques can be developed that support negotiators in becoming more adaptable (e.g., learning about cues and knowing how to respond to them).
We hope our findings will motivate further research to explore the nuances of strategic adaptability and inform the development of practical interventions to enhance negotiation effectiveness.
Notes 1.The inductive procedure was performed after analyzing study 2 and resulted in two additional categories.
2. Indeed, 65% of the participants used most adaptations in phase 3 or equally frequently to another phase.
3. We are aware that the participants functioned in dyads, however, the individual participants provided interesting insights on how strategic adaptability influences the outcome.

Table 2 .
The participants were 68 undergraduates following a purchasing and supply management course.Data were collected during an online one-day negotiation training.

Table 3 .
Notes: a Outcome group: Low-low including PM < 760, FM < 800; pareto-optimum PM ¼ 760, FM ¼ 800; low-low outcome group where the focal participant and the opponent both scored low;a Pareto-optimum group where the focal participant and the opponent both attained the maximum joint outcome; a low-high outcome group where the focal participant scored low but the opponent high; and a high-low outcome group where the focal participant scored high and the opponent low.
low-high PM < 760 or FM < 800; high-low PM > 760 or FM > 800; b participants can have similar amounts of adaptations in different phases and, therefore, they do not add up to 100%; c numbers are representing participants Source: Authors' own work a