Mind on the battlefield: what can cognitive science add to the military lessons-learned process?

ABSTRACT

In this paper, we demonstrate the advantages of combining the detailed military history of actual battles with contemporary decision-making cognitive theories. Specifically, we analyse two battle decisions made in Israeli-Arab wars, and interpret, through a cognitive lens, some of the decisions that were subsequently deemed “incomprehensible” in the civilian and military literature. This perspective permits analysis of multiple processes – such as risk evaluation, mental completion of unknown information, and estimation of enemy response. Moreover, it predicts that even highly experienced commanders are prone to the effect of cognitive biases, and may therefore make bad decisions with disastrous consequences.

KEYWORDS:

• Tactical decision-making
• cognitive biases
• military history
• imaginability bias
• overconfidence bias
• military lessons-learned

Introduction

The ability of military commanders to successfully execute a tactical decision-making process is known as estimation of the situation, and was one of the cornerstones of Frederick the Great’s military doctrine (Michel 1990). A century later, estimation of the situation became an integral part of official US Army doctrine, and has formed the basis of the common military decision-making process (MDMP) in most modern militaries (Marr 2001). Importantly – as noted by Captain Roger S. Fitch in 1909—estimation of the situation is used to express “a logical process of thought, which, applied to a concrete tactical problem, enables one to arrive at a definite strategical or tactical decision.” Moreover, “an estimate of the situation is valueless unless it reaches its logical termination in a decision upon a suitable plan of action” (Fitch 1909, 1). Accordingly, this rational perspective of decision-making has figured prominently in the tradition of military tactical command and control (C2), and has not changed significantly in over a century.

However, the cognitive perspective of decision-making does not fully adopt rationality as a maintained hypothesis (Kahneman and Lovallo 1993). Moreover, within the military discourse itself the rational perspective has been called into question as an adequate framework for analysis of the modern tactical decision-making process (Cohen et al. 2000; Shoffner 2000). An improved understanding of why experienced as well as inexperienced commanders occasionally fail to interpret information correctly, to make predictions, generate alternatives or even track the battlefield and change current actions based on plan changes, may require additional approaches to the traditional MDMP perspective (Fallesen 1993). Notably, ignoring cognitive factors relevant to decision-making may impair a successful lessons-learned (LL) process – which in turn may undermine the necessary continual update of tactical commanders’ training programs. It has been suggested that lessons learned are disseminated not only within communities of military practice, but can pervade and affect the entire military system. Thus, reliable analysis of tactical decision-making in general, and failure decision-making in particular, is of the utmost importance (Dyson 2019).

To improve lessons-learned capability within the military organization, various types of data-collection (followed by analysis, dissemination, and archiving) are employed (Waliński 2015). Based on the information gathered, the sequence of events is reconstructed (Jabłoński and Lis 2012). To this end, after-action reports and reviews (AAR – a.k.a. “debriefs”), in-depth individual and group interviews, combined with an examination of relevant documents (such as operation orders, operation logs, aerial photographs), are regularly used. Nevertheless, some tactical decision-making processes cannot be simplistically reconstructed in hindsight based solely on the information collected (Waliński 2015). The ongoing fruitful cognitive research program on decision-making may add further insights of after-action inquiry for the military domain. Specifically, various cognitive perspectives have been applied to risky decision-making in extreme battlefield contexts – including heuristics and biases (HB), fast and frugal heuristics (FFH), and naturalistic decision-making (NDM) (for a review, see Ning et al. 2020;‏ Klein 2015; Kahneman and Klein 2009). HB promoters focus on flaws in human cognitive performance, and have based their discussions on well-controlled lab experiments, whereas exponents of FFH and NDM tend to underline successful intuitive performance. An additional perspective highlights the complexity of the battlefield, which requires a problem-solving type of comparison of fragmented course of actions, rather than decision-making per se (for a review, see Brehmer 2000; Brehmer and Thunholm 2011; Thunholm 2011).

The role played by cognitive biases in strategic and political decision-making is widely discussed in the literature (e.g. Bar-Joseph and Yossef 2014; Vertzberger 1998). However, much less attention has been paid to the effect of biases on tactical events on the battlefield itself – mainly because the restructuring of specific events on the battlefield is much less common in civilian research, due both to the difficulties in collecting the data and debriefing the participants, and to the military’s initial objection to independent external investigation (Milstein 2011). In this article, we demonstrate a possible integration between military and cognitive research through two detailed studies about well-known battles from the ongoing Israeli-Arab conflict – namely, the Yehiam Convoy Battle (1948 Arab-Israeli War), and the Battle of the Chinese Farm (October 1973 War). In these two operations, the decision-making ran counter to the information available to the commanders. The results were disastrous for fighters and commanders alike, and in military and civilian research the decision-making was subsequently deemed “incomprehensible”. Accordingly, the lessons that could have been learned from these events were impaired (Milstein 1999). The accessibility of comprehensive research materials; the magnitude of the military campaigns in question; the battles’ significance in relation to the overall campaign; and the commanders’ proven expertise were the determining factors in selecting these two battles. The Arab-Israeli War of 1948 and the October 1973 War were both multifaceted conflicts that played out across numerous battlefields. The battles that occurred in these wars – such as the Yehiam Convoy Battle, and the Battle of the Chinese Farm—had a profound impact on the discourse that subsequently evolved into the Israeli ethos of command and military culture. Both battles were led by experienced and highly regarded commanders, and viewed as landmark events in Israeli military literature. Moreover, both became the subject of extensive and publicly accessible research.

In the following section, we briefly describe the main cognitive factors relevant to tactical decision-making, as suggested by the current cognitive literature (see Table 1). Next, we present the historical background and chain of events of both battles in question, as gleaned from the military data approved by military censorship for public viewing, and from multiple debriefings and documents collected by civilian scholars.

Table 1. Main cognitive factors relevant to tactical decision-making.

Cognitive factors	Definition in the Military Context
lmaginability bias	The tendency to evaluate the risks of future states of affairs (e.g.military operations) according to the ease with which these risks are imagined.
Availability bias	The tendency to base decisions on available intelligence or previous experience in the battlefield even if not compatible with the actual states of affairs.
Underestimation of outgroup bias	The tendency to underestimate the enemy (e.g.attribution of limited firepower or poor military capability).
Overconfidence bias (ingroup bias)	The tendency to be overconfident in group military capabilities.
Overconfidence bias (general)	The tendency to overestimate the ability to predict and control future outcomes (e.g. of battle, operations).
Higher-level prioritization bias	The tendency to prioritize intelligence according to the military rank of the informant rather than accessible, reliable, up-to-date data.
Optimism bias	The tendency to expect positive future states of affairs (e.g.victory in battle) despite the low probability based on given reliable intelligence.
Negative rejection bias	The tendency to avoid generating negative future states of affairs (e.g.total defeat).
Decision avoidance behavior	The tendency to avoid making a choice by postponing it or by continuing in the current course of action (e.g. operational plan) without change.
Prioritizing immediate action (delay discounting)	A behavioral bias toward immediate military action, overcoming the priority for potential better but later consequences.
Vulnerability to metacognitive experiences	Cognitive difficulty involved in changing perspective, reexamination of intelligence or considering alternative operational plans.
Over-emphasis on accountability	The tendency to shift from risk-aversion to risk-preference due to over-emphas is on accountability.

The analyses of the decision-making processes comprise several steps. First, we rule out the initial reasons for a failed decision as proposed by NDM – that is, lack of experience, and lack of information. We then inspect alternative explanations for some of the decisions, from cognitive theories – namely, several well-studied cognitive biases. Finally, we examine several, less common, factors in military decision-making literature that may have significant impact on decision makers – such as the effect of military rank of the information source; decision avoidance by adhering to past decisions; and the difficulty of suppressing previous embodied experiences, even though they occurred in different circumstances.

Cognitive bias on the battlefield

When relevant information is not instantly accessible, the likelihood of events cannot be reliably estimated; circumstances are marked by stress and time constraints; and decision-making processes become prone to systematic errors. These errors are known as cognitive biases. Often, these are the common conditions on the battlefield, when actions and risky decision-making are required. In the military tradition – and in some cognitive perspectives – expertise should significantly reduce such biases. Specifically, if experts have rich repertoires of patterns of models based on previous experience in their area of expertise, the mental models they generate and simulate during the decision-making process are expected to reflect their expertise (Klein 2015; Lipshitz and Ben Shaul 1997).¹ However, military field commanders often face unique and unfamiliar circumstances that defy any analogous experiences (Shortland, Alison, and Barrett-Pink 2018; Whiteman 1998). Moreover, the decision-making by military commanders in the battlefield is not limited to a single action but involves a complex chain of decisions reflecting ongoing battlefield dynamics (Brehmer 1992; Edwards 1962; von Moltke, 1871/1993). Thus, the battlefield context may require a commander to fill in unknown aspects of the current state of affairs (such as the terrain, the enemy’s capabilities, or its deployment), and to decide a course of possible actions, and possible outcomes for the ingroup (i.e. his own forces) or the outgroup (i.e. the enemy’s forces). Since these cognitively demanding tasks are easily susceptible to manipulations by multiple psychological and circumstance-dependent factors, an expert commander is also liable to be influenced by cognitive biases in his decision-making.

Several elements have been proposed as key factors in the appearance of cognitive biases during decision-making – some of which, in our view, may be highly relevant to tactical decision-making (see Table 1). There is the tendency to assess the risks of future conditions according to how easy they are to imagine – even when the evaluations do not reflect the actual likelihood of the risks described by Tversky and Kahneman (1974) as imaginability bias. This bias predicts that vividly portrayed risks may appear to be extremely dangerous. Conversely, future obstacles and risks may be grossly underestimated if they are either difficult to conceive, or simply do not come to mind (Sherman et al. 1985). In other words, decision-making errors are prompted by easily generated alternative scenarios (Ceschi et al. 2019).² Due to the vagueness of the information (especially concerning the enemy’s intentions) and feedback delay during warfare, a front-line commander may initially be deprived of relevant information about the state of battle (Brehmer 2000). Therefore, in the absence of real-time information, the commander’s decisions are based on alternative scenarios that reflect predictions of possible outcomes. In other words, at many time-points during dynamic decision-making in warfare, the commander relies on the quality and accuracy of his/her scenarios (Brehmer and Thunholm 2011; Williams 2010). Less accessibility to relevant data about actual battlefield conditions entails greater dependency on alternative scenarios, and may therefore make the decision-maker more prone to imaginability bias. Importantly, the generation, manipulation, and maintenance of alternative scenarios have been found to be most fragile under time pressure, which may reduce a decision-maker’s ability to generate alternative hypotheses (Alison et al. 2013). Time pressure often characterizes battlefield conditions, so we suggest including imaginability bias in the possible threat-list to decision-making on the battlefield that needs to be considered.

Tversky and Kahneman (1973) also linked between the imaginability aspect of an event and its availability – in other words, the frequency of one’s previous encounters with a given phenomenon or event determines the ability to imagine it. Thus, likely occurrences are easier to imagine than unlikely ones. In the military context, a field commander may tend to use the availability heuristic whenever he or she estimates the probability of an event by the ease with which it can be generated in their mind, based on past experience. In that regard, experienced commanders may be more prone to availability bias than inexperienced ones. Although the research on cognitive biases and the role they play during decision-making process has triggered much controversy since Tversky and Kahneman’s seminal papers, it also has greatly expanded.³ The identification of several specific cognitive biases may be found to be extremely relevant to the military domain.

Overconfidence in ingroup bias is manifested in an exaggerated estimate of the ingroup’s abilities relative to outgroups. In a military context, it has been suggested that organizational and cultural factors that encourage decision-makers’ strong beliefs in their own abilities for “beating the odds” may result in risk-seeking behaviour (Haerem et al. 2011).⁴ Furthermore, overconfidence bias may play a significant role in the blinding effect—particularly with regard to obstacles, problems, or errors (Shipman and Mumford 2011). Overconfidence can manifest itself not only in the limited context of a given task, but in a broader context, as well. In other words, strong beliefs in ingroup superiority (e.g. of one’s national troops) was found to be strongly associated with underestimation of outgroup capabilities (e.g. the enemy’s military), which may lead to failed decision-making (Matthews et al. 2018; for an extended review of overconfidence bias from a large-scale war perspective, see; Johnson 2004).

Another class of biases is the affect heuristic – namely, the influence of emotion evaluation on the decision-making process (Slovic et al. 2007). The contemporary cognitive view of emotions is that the primary function of emotions is to identify risks and benefits associated with one’s wellbeing and are relevant to one’s survival (Lang and Bradley 2010). Therefore, emotions pertain not only to how one thinks events may affect oneself, but also to how one might handle these events (Frijda 1988). This may apply not only to actual scenarios but to alternative ones, as well (Milshtein and Henik 2020; Milshtein, Hochman, and Henik 2020). One of the most well-known and extensively studied affective biases that has found to be consistent and robust is the optimism bias (Sharot 2011; Weinstein 1980) – namely, the unrealistic tendency to expect good things in the future. In the military context, investigating the optimistic bias has been devoted mainly to decisions at the strategic and national level (e.g. military invasions, or the launching of all-out or limited wars) (Nicholson 2013; for a recent review see; Vennesson and Huan 2018).⁵ Additional perspective to optimism bias is a tendency to reject negative future outcomes – that is, not only a tendency to be overconfident in a positive outcome, but also to reject the feasibility of negative outcomes, especially those involving oneself (Milshtein and Henik 2020; Reece and Matthews 1993). Decision-making in the military domain is often described as being high-risk with possible negative consequences – both for the commander himself, and for his troops. In such cases, the effect of optimism bias, and the tendency to reject negativity, can be catastrophic.

In a battlefield context, several specific factors that may have received less attention in the empirical psychology domain can be considered as a potential threat to the decision-making process: higher-level prioritization bias; over-emphasis on accountability; decision avoidance tendency; and vulnerability to metacognitive experiences. Higher-level prioritization bias in battlefield conditions is defined as the tendency of a field commander to prioritize the intelligence provided by higher-level command over that provided by field sources – even when the latter is from reliable sources. This tendency has several possible causes. First, in the military hierarchy, it is assumed that superior officers keep their subordinates informed about what is happening (Feld 1959). Second, the ability to adopt a contrary stand to that of senior commanders is limited due to passive dependence on authority that is typical of the relationship between ranks in the military (Foucault 1977; Freeman 1948). Moreover, higher-level prioritization bias can also be conceptualized as a variation of the imaginability bias: the notion that the High Command is ignorant of the true state of the battlefield and is capable of sending a field commander and his troops on what in reality is a suicide mission may be unfathomable.

Risk-sensitivity theory predicts that decision-makers shift from risk-aversion to risk-preference in times of need (Mishra 2014). In a battlefield context, when the need to win is amplified by over-emphasis on accountability, the decision-maker may be inclined to adopt risk-preference behaviour (Alison, Eyre, and Humann 2011). This bias may also be conceptualized as a specific variation of framing effects – goal framing (Levin, Schneider, and Gaeth 1998) – whereby operation orders are not merely a straightforward set of instructions to be carried out, but an emotional manipulation that results in the operational goal being framed in terms of gain versus loss. In extreme wartime situations, when the ingroup’s very survival is at stake, framing possible negative consequences can lead to over-emphasis on accountability, and result in the field commanders’ increased proclivity to risk-taking. Significantly, even highly experienced and esteemed commanders are susceptible to this bias.

An additional factor to consider in the analysis of tactical decision-making is decision avoidance behaviour, which is described as a “flight from decision” (Anderson 2003). Interestingly, previous findings suggest that greater blame or regret may be incurred by an active action (such as making a controversial decision) than from inaction. Thus, in conditions of uncertainty, decision avoidance was found to be the preferred strategy (Eyre and Alison 2007; Ritov and Baron 1990; Shortland and Alison 2015). In this sense, refraining from updating an operative plan in light of new information, and sticking to the original plan can also be described as decision avoidance behaviour. Metacognitive experiences – such as the ease or difficulty with which new information is processed, or a familiar cognitive model of the enemy is replaced with an unfamiliar new one, or opposition from commanders or subordinates is handled – have also been suggested as possible influential factors during the decision-making process (Schwarz 2004; Schwarz et al. 2021).

Analysing the battlefield from a cognitive perspective: two case studies

In this section, we demonstrate the use of the cognitive perspective to analyse the tactical decision-making by two experienced Israeli commanders – Ben-Ami Pachter, the Israeli commander in the Yehiam Convoy Battle (27 March 1948, First Arab – Israeli War), and Uzi Ya’iri, Commanding Officer (CO) of the 35th Paratroopers Brigade and senior commander in the “Chinese Farm” Battle (16 October 1973, 1973 Arab-Israeli War). Both planned and executed faulty battle plans that ended in defeat, with multiple casualties, and both sparked fierce controversy over their decision-making – among researchers and in the Israeli popular media alike – although both were highly experienced and esteemed commanders.

Shortly after the outbreak of the 1948 (or First) Arab-Israeli War, Pachter was appointed Commander of the 21st Infantry Battalion of the Carmeli Brigade.⁶ Although most military commanders of the Jewish forces were not professional soldiers and specialized mainly in guerrilla warfare, some of them had gained considerable practical experience in planning and executing limited operations and accompanying supply convoys. It has been suggested that those with considerable practical experience in specific operations should be regarded as experts (Klein 2015), and in that regard, after four months of daily clashes with armed Arab forces and securing roads, Pachter may be considered as an expert in such operations. In October 1973, Col. Yairi had already completed two years as commander of the Israeli foremost special forces unit (the General Staff Reconnaissance Unit) and as deputy CO of the regular 35^th Paratroopers Brigade, so he, too, had considerable tactical experience. Neutral decision-making (NDM) suggests that there are three primary reasons for unsuccessful decisions: inexperience, lack of information, and the act of explaining away evidence of an impending problem (Klein 1999). Since inexperience is the least likely explanation for Pachter’s and Ya’iri’s faulty decision-making, the other two factors should be examined, based on the course of the battles and the decision-making they entailed.⁷

The Yehiam convoy battlea⁸

On 27 March 1948-four months after the outbreak of the 1948 (or First) Arab – Israeli War – a Jewish supply convoy on its way to Kibbutz Yehiam (an isolated Jewish settlement in the Western Galilee district) was attacked by an Arab force. Forty-six Jewish fighters – including their Battalion Commander Ben-Ami Pachter – were killed in the battle; their supplies were looted; and most of their trucks destroyed. This painful defeat not only severely damaged the Jewish armed forces in the Western Galilee but also undermined the possibility of including this region within the borders of the future Israeli state (Milstein 1996). In Israeli military literature, it is generally agreed that the defeat at this battle was due mainly to Pachter’s design and execution of an ill-advised operational plan (Gelber 2004; Morris 1987). However, his decision-making in this case remains unclear, especially in light of his previous rich experience in securing civil convoys in hostile territory.

On 29 November 1947, the UN General Assembly adopted the Partition Plan as Resolution 181 (II), whereby the British Mandate rule over Palestine would end, and the territory would be divided between two sovereign states-one Jewish, one Arab. The 1948 war broke out the next day. However, Resolution 181 (II) was implemented only six-month later. In March 1948, British troops still staffed the Palestinian police stations and military bases, and Jewish-Arab military hostilities were chiefly confined to a low-intensity conflict (LIC) in the form of limited-scale attacks (by both sides) and a struggle for control of roads, mainly in border regions. The Western Galilee district – extending from the Mediterranean coast to the mountains adjoining the Lebanese border – was one of those regions. Seventeen Arab and Druze villages and nine rural Jewish settlements were scattered throughout the district. To protect the Jewish settlements, the 21st Infantry Battalion of the Carmeli Brigade under the command of Ben-Ami Pachter was placed in the area. Unlike the Jewish troops, the local Arab military forces were not formally organized, but in early 1948, Syrian troops of Jaysh al-Inqadh al-Arabi (the Arab Salvation Army), under the command of Adib Shishakli, entered the Galilee (Milstein 1996).

In the weeks before the battle, several successful Jewish raids on nearby Arab villages and an Arab support convoy sparked great anger and desire for revenge among the local Arab villagers. Pachter was fully informed of the relevant intelligence reports (Carmel 1948; Eshel 1949) – including several specific warnings from the Jewish intelligence service, local Arab informers and British military personnel prior to the convoy’s departure – that the Syrian troops of Jaysh al-Inqadh al-Arabi were planning an ambush of the supply convoy to Yehiam, and inciting the Arab villagers to help them. Relevant military reports from similar operations in other battle zones were also available to Pachter, and he was familiar with the modus operandi of armed Arab villagers and their ability to muster large forces from nearby villages in a short time. Accordingly, one might expect that his suspicions would be aroused when, despite the convoy’s departure at 2pm, when Arab village streets were usually full, the first village they passed was suspiciously empty. This was an warning sign that Pachter knowingly ignored (Hareuveni 1981; IDF Archives 2021; War of Independence: Collected Files 1/21 1948). In other words, there is no reasonable basis to believe he was unaware of the enemy’s intentions.

The strategic decision to dispatch an urgent supply convoy to Yehiam had been made higher up the chain of command – but Pachter was expected to revise his operational plans if conditions changed. These expectations matched the nascent legacy of infantry tactical field commanders as developed in the early days of the IDF, which called for independent command and encouraged initiative and autonomous decision-making on the frontline (Ronen 1989). In addition, supply vehicles frequently reached isolated settlements under cover of night, so an immediate departure of the convoy was not a matter of “life or death.” In summary, Pachter appears to have knowingly decided to send the convoy when he did, despite the risk of confronting Arab forces who were superior in number, and despite having the time and autonomous command to plan and execute an alternative operation.

Did Pachter have a factual basis to believe that his troops could safely escort the convoy on March 27, despite the Arab operational plans? This was just one of a series of supply convoy escort operations that were planned by Pachter and successfully executed by the 21st Battalion troops. This suggests that foremost in his mind were mental models of previous, successful convoys that had been escorted by his forces (i.e. availability heuristics and imaginability biases). Moreover, Pachter explicitly expressed his confidence in the qualitative superiority of his Jewish fighters to the Arab forces (Ben-Ezer et al. 1981). High self-esteem of the Jewish national group coupled with very low regard for the local Arab opponents was common during the war in the Zionist print media and in the military discourse (Nachšon 2020; Višnitzer 2005). It has been suggested that generating and maintaining a mental model of self (i.e. of the in-group nation) and of others (i.e. the out-group nation) may play a key factor in conflict behaviour, including national conflicts (Hogg, Abrams, and Brewer 2017; Kaplowitz 1990). Specifically, the Zionist national representative model of their typical soldier as a fearless warrior was forged as a counterpoint to the images of both the helpless Orthodox Jew of the Diaspora and the incompetent indigenous Arab of Palestine models (Šapira 1997; Šapîrô 1984). Importantly, recent findings indicate that strong beliefs in national superiority are associated with poor decision-making in the military context (Matthews et al. 2018).

Thirty minutes after leaving, seven vehicles carrying personnel (i.e. 87 Jewish soldiers) and supplies were ambushed by Arabs. Despite injuries and a loss of command, the leading armoured combat vehicle was able to make it through and reach Kibbutz Yehiam. However, the path for the convoy’s remaining vehicles was partially blocked by one vehicle that had overturned and was burning. Pachter, who was in that vehicle, ordered a retreat. In the ensuing exchange of fire, he was killed, over half the soldiers in the trucks were burned and killed in battle, and only four of the thirty-nine soldiers on the rearguard bus survived the nine-hour shootout. Pachter’s decision-making was severely criticised even before the war ended. However, the after-action debriefing failed to explain it.

The battle of the Chinese farm: case study II⁹

The Battle of the Chinese Farm took place some 25 years after the Yehiam Convoy Battle, on the banks of the Suez Canal during the 1973 Arab-Israeli (a.k.a. the October) War. Unlike the 1948 war, the fighting this time was between professional armies rather than armed militias. This may make a significant difference between the two decision-making contexts in question. However, apart from the catastrophic outcome for the Israeli troops in both battles, there are several additional and notable similarities. In both battles, the Israeli soldiers tried to break a way through a road held by superior Arab forces. The Israeli commanders were considered excellent field commanders, who were fully apprised both of the relevant information. Importantly, the decision-making that preceded the Battle of the Chinese Farm – like that in the Yehiam Convoy case – remains a mystery (Bergman and Melzer 2003; Gelber 2016; Tzur 2010). Certainly, none of those who took part in the Battle of the Chinese Farm were able to explain why Colonel Uzi Yairi, the Commanding Officer (CO) of the 35^th Paratroopers Brigade, decided to lead his soldiers into a “real suicide operation” (radio network recordings of Southern Command HQ, 1973, as cited by Ben Dor 2010).

“The Chinese Farm” was the name given in Israeli operational plans to one of the planned crossing points of the Suez Canal. The terrain was a mix of swamps and deep sand, which meant that traffic (especially for heavy vehicles) was only possible on a few gravel roads (Michelson 2003). The Tirtur Route was one of them. On the first day of the war (i.e. October 6th), Egyptian army forces crossed the Suez Canal and occupied a strip of land along the canal that was divided between two Egyptian Armies. The Tirtur Route crossed the perimeter of an Egyptian brigade that included infantry trenches and well-prepared personal foxholes – re-opening it was deemed essential for the planned Israeli counteroffensive (Ben Dor 2010). This task was assigned on October 16 to Yairi and his troops after several futile attempts by the CO of Israel’s Southern Front to do so.¹⁰

By this stage of the fighting, information from diverse sources had been received by Israeli intelligence regarding the Egyptian forces’ deployment and combat order. Reports of the failed attempts to open Tirtur Route the night before the battle in question were also forwarded to the brigade’s headquarters.¹¹ However, there is disagreement over the extent to which the Israeli High Command was aware of the relevant information in real-time.¹² It is possible that Yairi was not given the relevant information by his superiors. Moreover, he was told to deploy his soldiers over a wide area to search, find and destroy small Egyptian commando squads (i.e. “tank hunters”), that were not considered a significant threat to his Paratroopers’ Brigade. In an open terrain, when the enemy is well dug in and has superior firepower, this is an unacceptable operational plan, and contrary to regular infantry practice in such circumstances (Gelber 2016; Tzur 2010). Thus, lack of or misleading information may have been a satisfactory explanation for Yairi’s decision-making – were it not for the fact that he had, in fact, also been given different and contradictory intelligence. Two days before the battle in question, aerial photographs were presented to him by Southern Command intelligence, that indicated a heavy line of Egyptian personal dugouts along the Tirtur Route. In addition, close to the H-hour (the planned time of attack), Yairi and his Operations Officer met commanders and troops of another paratrooper battalion that had just returned from failed attempts to open the route. Those paratroopers shared with them updated information from the battlefield – namely, the location of multiple Egyptians fighting positions and Sagger anti-tank guided missile positions ranged along the route. Moreover, they explicitly warned Yairi against executing his plan as originally conceived, and noted that many wounded and dead Israeli soldiers remained along the route (Ben Dor 2010; Ben-Ari Kapusta 2006; Bergman and Melzer 2003). This updated intelligence presumably sparked Yairi’s doubts about the planned operation, as evident from his radio communication with Gen. Haim Bar-Lev, the Southern Front CO (the IDF’s former Chief of Staff, brought out of retirement especially for the purpose), shortly before H-hour. However, he apparently backed down and did not insist on making the necessary changes based on the updated information – on the contrary, he stuck to the original plan as he had been ordered, word for word, and rejected his staff officers’ alternative plan (Tzur 2010). Moreover, he decided to risk advancing without a professional Forward Observer (FO),¹³ despite his awareness of the key role played by artillery support in infantry battles (Adan 1979; Ben Dor 2010; Gelber 2016; Tzur 2010). In addition, he was satisfied with a vague promise that he could count on tank support, despite being aware of previous failures of Israeli tanks to make it through the Tirtur Route (Ben Dor 2010).

Col. Yairi’s decision-making suggests that he may have placed more faith in information from his superiors than in intelligence from frontline commanders (i.e. higher-level prioritization bias). Moreover, subordinates’ trust in the system that the superior commander represents may strengthen their personal trust in the commander himself (Shamir and Lapidot 2003) – and trust of Israeli society in the IDF and in its commanders was at its peak after the Israeli victory of 1967 (the Milstein 2011, 2019). In particular, Yairi received his orders directly from the most senior commander in the southern battlefields sector (i.e. Gen. Bar-Lev), whom he had served under in the past, and greatly admired.¹⁴ This may explain why Yairi ignored the explicit warnings he received. Nonetheless, it is unlikely that Yairi thought the updated intelligence to be false or irrelevant, given that he knew his sources and respected them as military professionals.¹⁵ Moreover, IDF military doctrine encouraged frontline commanders to take initiative and to make decisions in accordance with the current battlefield situation (Ben-Shalom and Shamir 2011). The notion that the Israeli High Command was absolutely ignorant of the actual battlefield situation, and therefore might be sending him and his troops on a doomed mission, may have been unfathomable to Yairi – a type of negative rejection bias.

Negative rejection bias—or the tendency to downplay or ignore negative evidence – may also be apparent in other aspects of Yairi’s decision-making. The updated intelligence presented to Yairi included explicit descriptions of killed or wounded Israeli soldiers who had been left along the route (Ben Dor 2010; Bergman and Melzer 2003) – but his decision-making did not indicate that he expected to fail, as well. Indeed, his actions suggest that he was confident of his troops’ ability to win the battle without aerial, armoured and artillery support. This may be a sign of overconfidence bias or underestimation of the Egyptian military capability. Underestimation of the Arab infantry and overconfidence were very common throughout the Israeli military and civil discourse before the October 1973 war (Ben-Zedeff 2017; Biniamini 1969, 1980; Gelber 2021). In the context of the 1973 Arab-Israeli War, the failure to re-open the Tirtur Route before the attempt by Yairi’s was attributed not to the training and equipment of the Egyptian infantry soldiers (as believe today), but rather to random Egyptian suicide attacks (Ben Dor 2010).

Yairi’s optimistic expectations may have been bolstered by his previous successes. Under his command, the 35^th Brigade had successfully pursued small Egyptian commando units in the southern Sinai in the first days of the war. However, the updated intelligence specifically stated that there were no similarities between those encounters and the conditions along the Tirtur Route – and this optimism bias may have caused Yairi to fail to match the situation with the appropriate action (Lipshitz et al. 2001). Cognitive flexibility or adaptiveness (i.e. metacognitive experience) may also determine the readiness with which a decision-maker processes new information. This may also bear upon the imaginability bias—namely, that processing new and contradictory information, and replacing one operational plan with another, may have posed a significant challenge for Yairi, as it required him to mentally generate multiple alternative scenarios that he may have had trouble imagining from the outset. The final factor in play was the pressure exerted upon him by both his superiors and his subordinates for immediate action. The paratroopers who remained with Yairi (i.e. only the 890^th Battalion) had not engaged in any significant battles, and were worried they might “miss” the war (Ben Dor 2010). In addition, Yairi’s commanders emphasized to him that he played a critical role in the Suez Canal crossing operation, which was already underway under the command of Maj. Gen. Ariel Sharon (Gelber 2016; Tzur 2010). In light of all the above, changing operational plans at this stage, while defying his superiors and subordinates, may have been too much pressure for Yairi to withstand. Thus, Yairi’s behaviour may best be characterised as change decision avoidance.

Accordingly, at 1:00pm, his troops set out to their destination. The three companies of the 890th Battalion were spread out under the command of Yitzhak Mordechai, followed by the Col. Yairi’s brigade command, and its staff. The troops made contact with the Egyptian brigade two hours after H-hour – where they were met with machine-gun fire, direct anti-aircraft fire, tank and mortar shells, were all directed at them. Most were killed or injured, as the battle developed into a desperate effort to evacuate the wounded – resulting in further casualties and injuries. All told, over a hundred paratroopers of the 890th Battalion were injured, and 43 killed. Ten armoured tank crew members were killed, twenty were injured, and four were missing. After 17 hours of fighting, the withdrawal was completed. Unlike Pachter in 1948, Yairi survived the battle, and took full responsibility for his decision-making – but he never properly explained why he rejected the information presented to him and accepted his superiors’ order without an argument (Ben Dor 2010).

Summary

In the military domain, failed decisions come at a heavy cost in human life, and have long-term political and social impacts. This in itself should be sufficient grounds for a special research effort into understanding why skilled field commanders, with relevant intelligence information, often make misguided and unacceptable tactical decisions. Nonetheless, applying insights from the cognitive research frontier to retroactive military and historical analysis is comparatively rare. In this paper, we demonstrated the potential of bringing together the explanatory power of cognitive theories and detailed military research of specific battles. In keeping with the NDM perspective (Klein 1999), we first inspected whether the fateful decisions taken in two battles could be accounted for by any or all of three common causes: inexperience, lack of information, and explaining away evidence of an impending problem. Subsequently, we examined whether various other factors noted in decision-making literature may offer a satisfactory explanation – such as the imaginability bias, optimism bias, negative rejection bias, or in-group and out-group biases. In addition, we proposed several other factors that are less common in military decision-making literature – such as the influence of the military rank of a given information source, adherence to previous decisions as a form of decision avoidance, and the difficulty to suppress previous embodied experiences, even when they occurred in different circumstances.

In the two battles in question, Israeli troops were led by experienced commanders – Pachter, who had just completed four months of intense and successful combat, and Yairi, a senior and highly experienced commander of an elite IDF unit. Moreover, both commanders were given updated information on the enemy’s intentions and its strength; in Pachter’s case, intelligence from behind enemy lines, and in Yairi’s case, reports from the battlefield. This information contradicted the intelligence assessments behind the orders that both commanders had received. Significantly, in both cases, the commander apparently assessed the updated information as reliable: in response, Pachter avoided taking the hills that dominated the road, to avoid encounters with the Arab fighters, while Yairi reported the updated intelligence to his superiors shortly before H-hour. One might expect, therefore, that they would have reassessed their feasibility of success as low, and change their battle plans accordingly – but they did not. Neither Pachter nor Yairi considered alternatives, nor tried to challenge their superiors’ order, or to persuade them of the benefit of adjusting the operation plans to the updated information; instead, they chose to stick with the original plans. Assuming the above accounts accurately described the events as they unfolded, a different perspective may be required to clarify the decision-making in question. We therefore suggest considering these instances through a cognitive lens, that takes into account not only explicit factors (e.g. the intelligence quality, unexpected behaviour of the enemy or the decision-maker’s experience) but implicit ones, as well – namely, alternative scenarios, generated, manipulated and maintained by the decision-makers, coupled with cognitive biases, which may shed light on the decisions in question.

Consideration of approximate outcomes is a preliminary stage in determining the success of a given military operation. Expecting positive outcomes (such as defeating the enemy at a low cost) can play a crucial role in deciding whether to engage in a battle, or avoid it. Expecting victory in defiance of low probability based on reliable intelligence, may be regarded as a form of optimism bias (Weinstein 1980), and there are several factors that may have bolstered that bias in the decision-making of the two commanders in question. First, shortly before the battles in question, Pachter and Yairi had experienced easy victories over seemingly similar enemy forces (i.e. armed Arab villagers in 1948, and small Egyptian commandos in 1973, respectively), and this availability heuristic may have biased their estimations of future success as being more probable. In other words, mental models of successful battles were more accessible to them, and easier to generate in their minds, than scenarios of military failure. Thus, although their updated intelligence indicated a much larger enemy battle order (i.e. the strategic arrangement of fighters and weapons in the battle) than in their previous encounters, they may have grossly underestimated that risk if they found it difficult to conceive certain possible dangers (i.e. imaginability bias). Underestimation of the Arabs in general, and Arab fighters specifically, was deeply rooted in Israeli culture, both in 1948 and 1973. Although Yairi did not expressly voice such disdain (as Pachter did), he presumably held a similar evaluation of his Egyptian infantry opponents – although the Arab militia fighters in March 1948 had scored many successes in the battle on the roads, and, the Egyptian soldiers October 1973 had also proven that they were well trained and capable of fighting. Conversely, overestimating the Israeli military capability may have prevented both from imagining outcomes of total Israeli defeat, despite the adverse conditions of the battlefield at that stage (i.e. negative rejection bias). In short, both Yairi and Pachter may have been encouraged to imagine positive future events rather than negative ones, in light of their recent victories (availability heuristics) and underestimation of their enemy/overestimation of their troops, which would have entailed difficulties in imagining, manipulating, and maintaining a Israeli defeat scenario (the imaginability bias).

An incorrect perception of the actual conditions in the battlefield underpinned the operation orders received by Pachter and Yairi. As a result, they were effectively sent by their superiors on suicide missions. We suggest that the difficulty in imagining such alternative scenarios as one of the factors that prompted both Pachter and Yairi to carry out their commands unquestioningly. This is a reflection of the implicit trust that Israeli soldiers have in their commanders, and in the IDF as a whole (Shamir and Lapidot 2003), and of the prevailing assumption throughout the military hierarchy that one’s superior officers are privileged to better information than oneself about what is happening (Feld 1959). Moreover, Yairi had a special relationship with his superior (Bar-Lev), which made this task even harder for him.

In military decision-making, it is not only the operation’s outcomes that need to be considered, but the consequences of not executing – or even refusing – an order, as well. The evidence suggests that the alternative of cancelling or postponing the planned operation was not an option for either Pachter or Yairi. In both cases, the operation was described as critical to the success of the entire campaign: Pachter’s mission was to save a besieged Jewish settlement, and Yairi may have thought that the entire operation of crossing the Suez Canal depended on him.¹⁶ This may not have helped them to consider all alternatives, and suggests that over-emphasis on accountability impaired their discretion (Alison, Eyre, and Humann 2011). Moreover, their own subordinates were eager to go into battle. Yairi’s troops, in particular, chafed against the idleness that had been imposed on them, and Yairi could well imagine them objecting strongly to a cancellation of the operation. This overestimation of the risks of alternative scenarios may be characterized as a case of imaginability bias.

After-action debriefing and systematic analyses of warfare usually focuses on the actual state of the battlefield – namely, what was known to those involved in the fighting, the extent to which they were prepared to confront the enemy, and whether or not unexpected and unpredictable events could in fact have occurred. However, human behaviour in general – and in extreme conditions such as combat, in particular – is not affected by explicit factors alone, i.e. the actual situation on the ground. Alternative scenarios, coupled with cognitive biases, may also play a significant role, and should therefore be considered in decision-making analysis. In a battlefield context, this means mentally completing partial intelligence; extrapolating possible outcomes to various courses of action; or manipulating alternative enemy initiatives. In other words, a host of cognitive biases can affect combat commanders’ decision-making, even when they have extensive experience and relevant intelligence available to them.

We suggest considering failed tactical decision-making in the battlefield that cannot be accounted for by explicit factors (such as an inexperienced commander, untrained soldiers, inadequate intelligence, or an unpredictable enemy) as possible evidence of cognitive factors. Two such cases were discussed in the present paper. This perspective may both inform commanders training programs, and advance empirical studies. In the military domain, the cognitive perspective can make a significant contribution mainly because it is clear that even experienced and senior commanders may be susceptible to cognitive biases and make wrong decisions, at a heavy price. These points should be taken into account when designing professional military education (PME). Simply stated, future PME programs for military personnel who are expected to confront the rigorous intellectual demands of major conflicts and complex contingencies may benefit greatly from the insights obtained in cognitive research. In the decision-making research domain, inspecting actual after-action debriefings and military-historical studies may provide researchers with greater insight into how risk decision-making in extreme conditions works in the field.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The supporting data for the findings of this study is available on request from the corresponding author, Dalit Milshtein.

Additional information

Notes on contributors

Dalit Milshtein

Dr. Dalit Milshtein is a neurocognitive researcher. She received her PhD from the School of Brain Sciences and Cognition at Ben-Gurion University of the Negev in Israel. Her research interests include decision-making, mental manipulation, collective experience, and social dynamics. She previously worked in the Social and Affective Neuroscience Lab at Haifa University and is currently affiliated with the Social Intelligence Group at Humboldt University Berlin.

Avishai Henik

Dr. Avishai Henik is a Distinguished Professor Emeritus of Psychology at Ben-Gurion University of the Negev. He has over 300 publications and studies attention, emotion and numerical cognition. Dr. Henik has been awarded a prestigious European Research Council (ERC) Advanced Grant to continue his cutting-edge research on numerical cognition. In addition, he won a Humboldt Research Award, in recognition of accomplishments in research and teaching, and the FENS-Kavli Network of Excellence Mentoring Prize 2020, for demonstrated leadership in fostering the careers of neuroscientists.

Eviathar H. Ben-Zedeff

Eviathar H. Ben-Zedeff is a research fellow in the International Policy Institute for Counter-Terrorism (ICT) in the Reichman University, Herzliya, Israel, and joint-coordinator of the Alfredo Workshops for Defence. He was the editor of the Israeli Defence Forces’ Ma’arachoth professional magazine. He holds a BA degree in History and International Relations from The Hebrew University at Jerusalem and an MA in Journalism from The University of Texas at Austin. His research focuses on Israeli military history, terrorism-press relations and military- press relations.

Uri Milstein

Dr. Uri Milstein is a philosopher of war, an independent war researcher and an expert in after-action debriefing. He fought in the IDF’s) Israeli Defence Forces (Paratroopers Brigade and the reserves, where he served as a soldier and combat medic on the front lines until the first Lebanon War in 1982. In addition, he served as an investigator and historian of the Israeli Paratroopers Brigade. He authored a four-volume history of IDF paratroopers, as well as a four-volume history of the beginning of the1948 Israeli-Arab war in Hebrew and English. He is currently working on a thorough study of the 1948 Israeli-Arab war and the foundation of the State of Israel that will include thirty books, eleven already published. He has authored scores of papers and books about significant battles in all Israeli wars. He developed and published a security philosophy based on the “principle of survival” in both Hebrew and English. He often publishes detailed military debriefings in the Israeli press.

Notes

1. The question of how much more information benefits decision-making efficiency is also debatable. For example, it has been suggested that under ecological conditions (rather than controlled laboratory ones), experts’ decision-making can be improved, and judgements made more accurate, by actually ignoring information (Gigerenzer and Gaissmaier 2011). Similarly, previous studies found that although military decision-making uncertainty was reduced as information increased, it did not improve plan quality, or reduce planning time (Thunholm 2008).

2. By “alternative scenarios,” we mean mental structures or representations that are different from the present state of affairs, or from the information available to the decision-maker (for recently review, see the Milshtein and Henik 2024).

3. For example, it has been suggested that quick decisions are based on unconscious and associative processes that ignore parts of the information. These heuristic-based decisions or judgments may be highly economical and usually effective (Gigerenzer and Gaissmaier 2011; Kahneman 2011). On the other hand, heuristics may provide non-optimal solutions (Wójtowicz and Winkowski 2018). In other words, heuristic strategies for quick decisions could provide practical but somewhat unreliable decisions.

4. Recent findings suggest that reducing the problem-solving confidence of junior field commanders predicted better performance (Calleja, Hoggan, and Temby 2020).

5. The present Russian-Ukrainian war may provide another demonstration of the optimistic bias in wartime decision-making. The Russian government assumed it would be able to replicate in Ukraine the success they had gained in putting down the Syrian rebellion, despite being aware of the fundamental differences between the two battlefields. This bias persisted after their military failures in the first phase of the war. More generally, the failure of the optimistic Russian predictions of Ukrainian and NATO responses resulted in heavy Russian losses in terms of casualties, weapons, international relations, and even damage to President Vladimir Putin’s personal reputation within the intra-Russian political arena.

6. The formation of the IDF (Israeli Defence Forces) as a conventional military actually began shortly after the war had commenced.

7. Reconstructing a battlefield in general – and in the specific cases in question in particular – is not a simple matter, and involves great disagreements between researchers. Posing questions about commanders’ responsibility for events also triggers an emotionally charged discourse. Nevertheless, despite the differences between various accounts, we believe it is impossible to ignore the questions that arise with regard to the tactical decision-making.

8. Part of the source material for this vignette is based on personal communications with: Y. Avni; I. Ben-Barak; A. Bannai; M. Carmel; Y. Ezroni; S. S. Hanafas; Y. Pundack; Y. Reshef; M. Šacheviz; and R. Weitzreiber. For comprehensive review on the war see Kimche, and Kimche (1960); Lorch (2019); the Milstein (1996).

9. Part of the source material for this vignette is based on personal communications with: G. Kopel; G. David; Y. Mordechai; and Y. Barkan. For a comprehensive review of the war, see Aker (2014); Gawrych (2015).

10. Major-General Haim Bar-Lev, the IDF’s former Chief of Staff (CoS), 1968–71, and then a minister in the Israeli government. During the war he was urgently called back into active reserve service due to Israeli military failures on both northern and southern fronts (Milstein 1993).

11. On October 15, the Israeli 14th Brigade had tried unsuccessfully to break through the Tirtur Route. Survivors of the battle and the brigade’s armoured battalion soldiers reported encountering enormous numbers of Egyptian infantry fighters (14th Armored Brigade, IDF 1975; Division 162 reports, IDF, as cited in Ben Dor 2010; Reshef 2011)

12. In this matter see also The Agranat State Commission, January 28, (1975) (1975); Ben Dor (2010)

13. A Forward Observer is an officer responsible for artillery support direction and friendly fire prevention.

14. Yairi had been Bar-Lev’s chief of staff while the latter served as the IDF’s supreme commander (Ben Dor 2010).

15. For instance, one of his sources was Lt. Col. Michael Ben-Ari Kapusta – one of the founders of the General Staff Reconnaissance Unit, which Yairi later commanded), and a role model for Yairi’s generation of officers (the Milstein 1985, 2019).

16. As it happens, in both cases, these assumptions were false: Kibbutz Yehiam was not, in fact, under siege, and received supplies on a regular basis (the Milstein 1996(, and in 1973, while Yairi’s soldiers were being killed on the Tirtur Route, other troops managed to reach the Canal by a parallel route (Ben Dor 2010; Tzur 2010).